Introduction
The aim of primary prevention is to reduce the risk of first-ever stroke in asymptomatic people. A large part of stroke risk can be attributed to potentially modifiable risk factors [1, 2].
The INTERSTROKE study, a large case-control study including 13 447 cases of acute first-ever stroke recruited between 2007 and 2015 in 32 countries, identified 10 risk factors that were associated with 91% of population attributable risk (PAR): hypertension, current smoking, waist-to-hip ratio, diet risk score, physical activity, apolipoprotein (Apo) B/ApoA1 ratio, psychosocial factors, cardiac causes, alcohol consumption, and diabetes mellitus [1] (Figure 7.1).
Figure 7.1 Ten potentially modifiable risk factors associated with 90.7% of population attributable risk (PAR) for stroke, ischemic stroke, and intracerebral hemorrhages in the overall population in the INTERSTROKE study. (From O’Donnell et al. [1].)
Apo = apolipoprotein; mAHEI = modified Alternative Healthy Eating Index; OR = odds ratio; T = tertile.
* Composite PAR includes all 10 risk factors; self-reported history of hypertension or blood pressure ≥140/90 mmHg was used for hypertension variable.
The Global Burden of Disease (GBD) Study, a worldwide observational epidemiological study describing the mortality and morbidity from major diseases, injuries, and risk factors to health from 1990 to the present, attributed 90.5% of the global stroke burden (as measured in disability-adjusted life years [DALYs]) in 2013 to modifiable risk factors which included behavioral factors (smoking, poor diet, and low physical activity), metabolic factors (high systolic blood pressure [SBP], high BMI, high fasting plasma glucose, high total cholesterol, and low glomerular filtration rate), and environmental factors (air pollution and lead exposure) [2] (Table 7.1).
Table 7.1 Top risk factors ranked by number of Disability-Adjusted Life Years attributable to stroke for both sexes combined in 21 regions in the Global Burden of Disease Study 2013
 |
The strategy in primary prevention is to lower stroke risk attributed to modifiable factors through education, lifestyle changes, and medication. Non-modifiable risk factors include old age and some genetic factors. The negative influence of conditions such as arterial hypertension, atrial fibrillation, or diabetes mellitus can be lowered by controlling and treating the underlying disorder. Targets for stroke prevention can be directed either at the entire population or at high-risk individuals who are already suffering from disorders such as hypertension or diabetes mellitus. Usually, in general medicine, the latter approach is more prevalent. Therefore, issues relating to the high-risk approach of stroke prevention shall be the focus of this chapter.
Lifestyle Factors
Stroke prevalence has been associated with individual lifestyle factors in several studies. Healthy lifestyle in general was considered in one large prospective cohort study of healthy women. In this study, healthy lifestyle, consisting of abstinence from smoking, low to normal body mass index (BMI), moderate alcohol consumption, regular exercise, and healthy diet, was found to be associated with a reduction in ischemic stroke (relative risk [RR] 0.3; 95% confidence interval [CI] 0.1–0.6) [3]. Using the data of two large cohort studies, the Nurses’ Health Study (71 243 women) and the Health Professionals Follow-up Study (43 685 men), Chiuve et al. [4] defined a low-risk lifestyle score based on the five lifestyle components: non-smoking, moderate activity ≥30 minutes/day, healthy diet, BMI <25 kg/m2, and modest alcohol consumption (men 5–30 g/day, women 5–15 g/day). The total number of low-risk factors was associated with a significantly reduced risk of total and ischemic stroke in men and women. Persons with low-risk lifestyle (all five low-risk lifestyle factors) had a decreased risk of stroke compared to persons fulfilling none of the low-risk lifestyle factors, RR 0.2 (95% confidence interval [CI] 0.1–0.4) and RR 0.3 (95% CI 0.2–0.5) for women and men respectively. However, only 2% of women and 4% of men were at low risk for all five factors. A large Finnish prospective cohort study (36 686 participants, 1 478 stroke events) found that the number of healthy lifestyle indicators (smoking, BMI, physical activity, vegetable and alcohol consumption) is inversely associated with the risk of total, ischemic, and hemorrhagic stroke [5]. Similar results have been found in large Swedish and Chinese cohorts [6–8].
Lifestyle modifications have a high potential to prevent at low cost and low risk the development of stroke risk factors such as diabetes, dyslipidemia, obesity, and hypertension, and should be an important issue in stroke prevention. Internet-based and mobile technologies offer new, cost-effective, widely accessible, flexible and personalized alternatives for the assessment of stroke risk, risk factor management, and the facilitation of behavior modification [9, 10].
Five low-risk lifestyle factors with a high potential to prevent stroke are:
non-smoking
moderate activity ≥ 30 minutes/day
healthy diet
BMI <25 kg/m2
modest alcohol consumption.
Cigarette Smoking
Smoking is a well-documented preventable risk factor of stroke. Large observational studies have shown cigarette smoking to be an independent risk factor for stroke in both men and women, with current smokers having a 2- to 4-fold increased risk of stroke compared with non-smokers (Figure 7.2) [1, 11–13]. An individual participant meta-analysis based on 25 cohorts found an increased risk of stroke of hazard ratio (HR) 1.58 (95% CI 1.40–1.78) for current smokers and of 1.17 (1.07–1.26) for former smokers [14]. A dose-response relationship was identified with an increasing risk with daily cigarette consumption in current smokers (HR for linear trend per 10 cigarettes 1.25; 94% CI 1.19–1.31) [14].
Figure 7.2 Increasing risk of stroke associated with number of cigarettes smoked for all stroke, ischemic stroke, and intracerebral hemorrhagic stroke.
Stroke risk for smokers as compared to non-smokers differed between stroke types, being highest for subarachnoid hemorrhages (SAH) (odds ratio [OR] 2.9; 95% CI 2.5–3.5), nearly 2-fold for ischemic stroke (OR 1.9; 95% CI 1.7–2.2), and no clear relationship for intracerebral hemorrhages (ICHs) (OR 0.7; 95% CI 0.6–1.0) [15]. Smoking is a well-established risk factor for ischemic stroke [8]. A meta-analysis focusing only on SAH found a RR of 1.9 (95% CI 1.5–2.3) for two longitudinal studies and an odds ratio of 3.5 (95% CI 2.9–4.3) for seven case-control studies [16]. One meta-analysis studying the risk factors for ICH found an adjusted RR for current smokers of 1.3 (95% CI 1.1–1.6; 13 studies) and an adjusted RR of 1.1 (95% CI 0.9–1.3; 12 studies) for ever having smoked [17]. Another meta-analysis including three studies from Europe/North America and three from Asia found that the RRs associated with current smoking compared to never smoking for ICH were 1.77 and 1.09 for the two regions, respectively [18].
Non-smokers exposed to tobacco smoke were estimated to absorb only the equivalent of 0.1–1 cigarette based on urine nicotine. Nevertheless, passive smoking was associated with a greater progression in atherosclerosis [19]. Never smokers exposed to tobacco smoke had in the period of 3 years a mean increase of intimal-media thickness (IMT) of the carotid artery of 31.6 μm (SD ± 2.0) compared to 25.9 μm (SD ± 2.1; p = 0.010) for non-smokers not exposed to smoke. The mean increase of IMT for current smokers was 43.0 μm (SD ± 1.9) [19]. Only a few studies investigated stroke risk due to environmental tobacco smoke exposure. A meta-analysis of 16 studies of variable design and quality suggests that spousal cigarette smoking is associated with an increased stroke risk (RR 1.3; 95% CI 1.2–1.4). The relative risk of stroke found for the highest level of exposure was 1.6 (95% CI 1.3–1.8) [20].
Smoking may potentiate the effects of other risk factors. The results of a large cohort study suggests a lower tolerance for high BP, increased BMI, and blood glucose levels in smokers which may expose them to a higher risk of fatal stroke [21]. Similarly, an interaction between smoking and the use of oral contraceptives increasing the risk of stroke was noted for women [22].
Smoking cessation reduces stroke risk rapidly. In a meta-analysis the HR for a decrease in stroke risk since smoking cessation in former smokers was 0.87 (0.84–0.91) for a linear trend per 10 years [14]. In the Framingham Study stroke risk had decreased 5 years after quitting smoking to the level of non-smokers [11]. In the Nurses’ Health Study total and ischemic stroke excess risk nearly disappeared after 2 years and relative risk for former smokers compared to never smokers was 1.4 (95% CI 1.0–1.7), while it was 2.6 (95% CI 2.1–3.2) in current smokers [12]. In a pooled analysis of three large Japanese cohort studies age-adjusted risk for stroke mortality had declined 5–9 years after smoking cessation by 23% in women and by 10% in men, and after 10–15 years it was decreased by 35% in men and 49% in women and no longer differed from people who had never smoked [13].
The benefits of quitting smoking are evident; however, due to its addictive effect the success in smoking cessation is only modest. Several behavioral and pharmacological therapies are available to assist smokers in quitting and their effects are the subject of a number of Cochrane reviews (e.g. [23–26]). All forms of nicotine replacement therapy (nicotine gum, transdermal patches, nasal spray, inhalers, tablets) are effective in increasing abstinence from smoking (RR 1.6; 95% CI 1.5–1.7) [23]. The anti-depressants bupropion (RR 1.6; 95% CI 1.5–1.8) and nortriptyline (RR 2.0; 95% CI 1.5–2.8) are also successful for smoking cessation. Their effect seems, however, to be independent of their anti-depressant effect and they are of similar efficacy to nicotine replacements [24]. The nicotine receptor partial agonist varenicline was also found to increase the chance for 6 months or longer abstinence when compared with pharmacologically unassisted quit attempts (2.24; 95% CI 2.06–2.43), with bupropion (1.39; 95% CI 1.25–1.54) or with nicotin replacement therapy (1.25; 95% CI 1.14–1.37) [25]. Psychosocial interventions such as behavioral therapy, self-help, or telephone counseling are effective, but have to be intensive [26]. Interactive, tailored internet-based or mobile phone-based interventions are moderately more effective than control programs to quit smoking at 6 months [27, 28].
Stroke risk for smokers is 2.9-fold for subarachnoid hemorrhages and nearly 2-fold for ischemic stroke. Even passive smoking was associated with increased risk for stroke.
Alcohol Consumption
Excessive alcohol drinking increases all-cause mortality, as well as the risk of coronary heart disease and stroke. A meta-analysis including 27 prospective observational studies found for heavy drinking (more than four drinks/day) an increased risk of ischemic stroke (RR 1.14; 95% CI 1.02–1.28), intracerebral hemorrhages (RR 1.67; 95% CI 1.25–2.23), and subarachnoid hemorrhages (RR 1.82; 95% CI 1.18–2.82) [29].
The relationship between alcohol and overall as well as ischemic stroke risk was described as J-shaped. This suggests that benefits overcome the harmful effect of alcohol at light to moderate alcohol consumption. Light (less than 1 drink/day) and moderate (1–2 drinks/day) alcohol consumption were associated with a reduction in ischemic stroke (RR 0.90; 95% CI 0.85–0.95 for light and RR 0.92; 95% CI 0.87–0.97 for moderate alcohol consumption) [29]. No association between light and moderate alcohol consumption and the risk of intracerebral or subarachnoid hemorrhages were found [29].
The apparently positive effect of light to moderate alcohol consumption is still under discussion. Beneficial effects on lipids, hemostatic factors, insulin sensitivity, inflammatory markers, and flow-mediated vasodilatation have been reported. Especially the flavonoids of red wine have been presumed to be involved in preventing the formation of atherosclerotic plaques [30]. Comparing the type of alcoholic beverage consumed, wine seems to be associated with the lowest risk of stroke [31]. The pattern of drinking seems to influence the vascular risk; binge drinking, even when alcohol consumption was otherwise light, increases the risk of ischemic and total stroke [32]. It has been suggested that alcohol is a trigger for ischemic stroke onset, and that stroke risk is elevated 2- to 3-fold after alcohol abuse within 24 hours or within 1 week preceding stroke [33].
Heavy alcohol intake and binge drinking increase heart rate, blood pressure, and the risk of hypertension and atrial fibrillation, thereby increasing stroke risk; this seems to be especially true for hemorrhagic stroke [34, 35]. In hypertensive subjects stroke risk was increased significantly by heavy drinking. In a 26-year Japanese prospective cohort study hemorrhagic stroke risk (RR 3.1; 95% CI 1.1–9.1) and to a lesser extent ischemic stroke risk (RR 2.0; 95% CI 1.1–3.6) were increased significantly in hypertensive heavy drinkers compared to non-drinking and light-drinking hypertensive subjects, whereas for non-hypertensive persons the increased risks of hemorrhagic stroke (RR 1.7; 95% CI 0.6–4.9) and ischemic stroke (RR 1.4; 95% CI 0.8–2.5) attributed to heavy drinking were not significant [35]. Reducing alcohol intake was found to reduce BP in a dose-dependent manner with an apparent threshold effect at two drinks per day in a meta-analysis of 36 interventional studies [36]. Heavy long-term alcohol consumption (more than three drinks/day) and episodic heavy drinking increase the risk of atrial fibrillation, a major risk factor of stroke [37].
Excessive alcohol drinking increases all-cause mortality, as well as the risk of coronary heart disease and stroke, but benefits overcome the harmful effect at light to moderate alcohol consumption levels. Nevertheless, binge drinking is a trigger for ischemic stroke onset (stroke risk is elevated 2- to 3-fold after alcohol abuse within 24 hours).
Obesity
A high BMI (≥25 kg/m²) is associated with an increased risk of stroke in men [38, 39] and women [40, 41]. Ischemic stroke rate increases in a dose-dependent manner with BMI [38, 42]. The relationship between hemorrhagic stroke and BMI is less clear: some studies found no influence of BMI on hemorrhagic stroke risk [39, 40, 42], whereas others found an increased [38, 43] or even a decreased [41] risk of hemorrhagic stroke for people with elevated BMI.
Measures of abdominal adiposity (waist circumference, waist-to-hip ratio, or waist-to-height ratio) have been suggested to be better indicators for stroke risk than overall body mass (measured by BMI) [1, 42]. However, after adjusting for other risk factors, risk prediction models using waist-to-hip ratio, waist circumference, or BMI to assess obesity did not differ in a clinically significant way [44].
Obesity is associated with an increased risk of hypertension, diabetes, dyslipidemia, atrial fibrillation, and obstructive sleep apnea. Adjusting for these confounding risk factors often attenuates the effect of body mass [38–40, 43–45]. A pooled analysis of 97 prospective cohorts with 1.8 million participants (31 093 stroke events) showed that 76% of the excess risk of BMI on stroke is mediated by the metabolic factors blood pressure, cholesterol, and glucose [45]. The effect of the metabolic factors was larger (98%) in overweight individuals (BMI 25–29 kg/m²) than for obese individuals (BMI ≥30 kg/m²; 69%). These results suggest that health interventions that control blood pressure, cholesterol, and glucose can reduce to a great extent the excess risk of stroke in overweight, but not fully in obese, individuals – this is crucial because weight-loss interventions often only have small long-term success.
Weight reduction was associated with improvement in BP, insulin sensitivity, blood glucose level, triglyceride and high-density lipoprotein (HDL) concentration, and markers for inflammation. In a meta-analysis systolic BP was reduced by 4.4 mmHg and diastolic BP by 3.6 mmHg for an average weight loss of 5.1 kg [46]. However, the randomized controlled Look AHEAD (Action for Health in Diabetes) study in 145 overweight or obese patients with type 2 diabetes found no significant difference in cardiovascular or stroke events between a group with intensive lifestyle intervention that promoted weight loss through decreased caloric intake and increased physical activity and a group receiving diabetes support and education after a median follow-up time of 9.6 years [47].
Combined interventions including dietary and exercise strategies with cognitive-behavioral therapy were the most successful for weight loss [48]. Increasing the intensity of psychological intervention resulted in greater weight reduction [48].
Obesity (high body mass index or high waist-to-hip ratio) is associated with an increased risk of stroke. Seventy-six percent of the excess risk of BMI on stroke is mediated by the metabolic factors blood pressure, cholesterol, and glucose.
Physical Inactivity
Several prospective longitudinal population studies have shown the protective effect of regular physical activity for stroke and cardiovascular disease in women and men [49]. Physical activity and the incidence of stroke were found to be associated in a dose-response relationship. A meta-analysis suggests that stroke risk can be reduced by 18% by an inactive individual increasing his or her physical activity to 11.25 metabolic equivalents of task (METs) hours/week (corresponding to international recommendations of at least 150 minutes per week of moderate-intensity aerobic physical activity, or 75 minutes per week of vigorous physical activity) [50]. A similar relationship was found for physical activity and risk of ischemic stroke – the biggest reduction in relative risk is seen from low level to moderate amount of physical activity [51]. Only a few studies investigated the effect of physical activity on hemorrhagic stroke. In a meta-analysis of three cohort studies high activity significantly decreased hemorrhagic stroke risk when compared with low activity (RR 0.66; 95% CI 0.48–0.91) [52]. On the other hand, the results of two large Japanese and Chinese cohort studies suggest a J-shaped association between the physical activity level and the risk of hemorrhagic stroke, with moderate activity levels being optimal, suggesting that vigorous physical activity in individuals with uncontrolled hypertension may be harmful (Figure 7.3) [53, 54].
Figure 7.3 Dose-response relations between physical activity levels and risk of ischemic stroke events and other diseases: meta-analysis for the Global Burden of Disease Study 2013. Higher levels of total physical activity were associated with lower risk of all outcomes. Major gains occurred at lower levels of activity, and the decrease in risk was minimal at levels higher than 3 000–4 000 MET minutes/week. MET = metabolic equivalent (ratio of working to the resting metabolic rate; 1 MET is the amount of oxygen consumed while sitting quietly; 600 MET minutes/week is the minimum level recommended by WHO).
The definitions of activity levels and activity types vary considerably between studies and the amount of activity is generally self-assessed. Therefore, METs were introduced to make energy expenditure more comparable. One MET is defined as 1 kcal/kg and is approximately equivalent to the energy cost of sitting quietly. Study results about the optimal type of activity for stroke prevention are inconsistent. Only a few studies have evaluated the influence of occupational physical activity. A meta-analysis distinguishing between leisure and occupational physical activity found a protective effect of both activity types [55]: people active at work had a decreased risk for ischemic (RR 0.6; 95% CI 0.4–0.8) and hemorrhagic stroke (RR 0.3; 95% CI 0.1–0.8), as well as those physically active during leisure time (RR 0.8; 95% CI 0.7–0.9 for ischemic stroke; RR 0.7; 95% CI 0.6–1.0 for hemorrhagic stroke). Commuting physical activity (walking or cycling to work) may also reduce stroke risk [56].
A few randomized controlled trials (RCTs) investigated the effect of physical activity interventions on stroke incidence. Whereas two small RCTs found a significant reduction of cardiovascular or stroke events after a physical activity program [57, 58], the LIFE and Inter99 studies – two large RCTs – found no such effect [59, 60]. There is not enough evidence for the type and intensity of fitness training protecting best against stroke.
The favorable effect of physical activity is at least partly mediated through effects on other risk factors. Physical activity decreases body weight and BP, and has beneficial effects on blood lipids and glucose levels [48, 61]. Nevertheless, it has been suggested that physical activity cannot entirely counteract the increased cardiovascular risk associated with obesity and being overweight [62].
Regular physical activity has a protective effect for stroke, probably mediated through beneficial effects on other risk factors.
Dietary Factors
Poor dietary habits contribute to the development of other stroke risk factors such as obesity, diabetes, hypertension, and dyslipidemia. Changes in dietary habits therefore have high potential for reducing stroke risk. Different foods and nutrients have been suggested to influence stroke risk via several mechanisms, e.g. by influencing BP, insulin resistance, inflammation risk, platelet function, endothelial function, and oxidation (Tables 7.2 and 7.3 [63, 64]).
Table 7.2 Meta-analyses of prospective cohort studies of foods and the risk of stroke
| Food | Studies | Subjects | Events | Unit | RR | 95% CI |
|---|---|---|---|---|---|---|
| Beneficial | ||||||
| Chocolate13 | 5 | 116664 | 4260 | Highest vs lowest category | 0.79 | 0.70–0.87 |
| Fruits14 | 16 | 964142 | 46203 | Per 200 g (2 servings)/d | 0.82 | 0.74–0.90 |
| Vegetables14 | 13 | 441670 | 14973 | Per 200 g (2 servings)/d | 0.87 | 0.79–0.96 |
| Fish15 | 8 | 394958 | 16890 | ≥5 vs 1 serving/wk | 0.88 | 0.81–0.96 |
| 2–4 vs <1 serving/wk | 0.94 | 0.90–0.98 | ||||
| Milk16 | 14 | 603920 | 25269 | Per 200 g/d increment | 0.93 | 0.88–0.98 |
| 125 g/d milk intake | 0.86 | 0.82–0.89 | ||||
| Eggs17 | 7 | 308000 | 8889 | High (1 egg/d) vs low (<2/wk) | 0.88 | 0.81–0.97 |
| Tea18 | 8 | 307968 | 11329 | Each 1 serving/d (1 cup) | 0.94 | 0.90–0.97 |
| Coffee19 | 17 | 1283685 | 12030 | Highest vs lowest category | 0.95 | 0.84–1.07 |
| 2nd highest vs lowest category | 0.80 | 0.75–0.86 | ||||
| 3rd highest vs lowest category | 0.89 | 0.84–0.94 | ||||
| Neutral | ||||||
| Whole grains20 | 6 | 245012 | 2337 | Per 90 g/d (≤120–150 g/d) | 0.88 | 0.75–1.03 |
| Nuts21 | 11 | 396768 | 9272 | Per 28 g/d increment | 0.93 | 0.83–1.05 |
| High vs low | 0.89 | 0.82–0.97 | ||||
| Cheese16 | 8 | 282439 | 9919 | Per 40 g/d | 0.97 | 0.94–1.01 |
| Legumes22 | 6 | 254628 | 6690 | Each 4 servings/wk (400g) | 0.98 | 0.84–1.14 |
| Butter16 | 3 | 173853 | 5299 | Per 10 g/d increment | 1.00 | 0.99–1.01 |
| Harmful | ||||||
| Red meat processed23 | 17 | 2079236 | 21730 | >50 g (1 serving)/d | 1.14 | 1.05–1.24 |
| >0 g/d | 1.17 | 1.09–1.27 | ||||
Table 7.3 Meta-analyses of prospective cohort studies of nutrients and the risk of stroke
| Nutrient | Studies | Subjects | Events | Unit | RR | 95% CI |
|---|---|---|---|---|---|---|
| Beneficial | ||||||
| Dietary potassium4 | 16 | 639440 | 19522 | Highest vs lowest* | 0.87 | 0.80–0.94 |
| Omega-3 polyunsaturated fat5 | 14 | 514483 | 9065 | High vs low | 0.87 | 0.79–0.95 |
| Total dietary fiber6 | 8 | 277537 | 9931 | Per 7 g/d | 0.93 | 0.88–0.98 |
| Neutral | ||||||
| Monounsaturated fat7 | 10 | 314511 | 5827 | Higher intake | 0.86 | 0.74–1.00 |
| Omega-3 plant sources8 | 3 | 98410 | 1300 | High vs low | 0.96 | 0.78–1.17 |
| Calcium9 | 10 | 371495 | 10408 | Highest vs lowest quintile | 0.96† | 0.89–1.04 |
| Saturated fat10 | 12 | 339090 | 6226 | High vs low | 1.02 | 0.90–1.15 |
| Total transfat10 | 3 | 190284 | 1905 | High vs low | 1.07 | 0.88–1.28 |
| Glycemic index11 | 7 | 225205 | 3046 | High vs low | 1.10 | 0.99–1.21 |
| Total carbohydrate11 | 4 | 170348 | 1851 | High vs low | 1.12 | 0.93–1.35 |
| Harmful | ||||||
| Glycemic load11 | 6 | 222308 | 2951 | High vs low | 1.19 | 1.05–1.36 |
| Dietary sodium12 | 10 | 72878 | not stated | Higher intake vs low | 1.24 | 1.08–1.43 |
CI indicates confidence interval; and RR, risk ratio.
* Median potassium intake was 103.0 mmol/d (range 68–149.8) in the highest category and 50.1 mmol/d (24–100.1) in the lowest one. The pooled RR was lowest at 90 mmol (≈3500 mg)/d of potassium daily intake (RRs, 0.78; 95% CI, 0.70–0.86).
† High dairy calcium intake was significantly associated with an ≈24% reduction in the risk of total stroke (RR, 0.76; 95% CI, 0.66–0.86), whereas non-dairy calcium intake showed no effect (RR, 1.01; 95% CI, 0.82–1.24).
Fruits, Vegetables, and Whole Grain
In large epidemiological studies, high fruit and vegetable intake were associated with a decreased risk of stroke. A meta-analysis including 25 cohort studies calculated a 13–18% reduction in the RR of stroke, for each 200 g/day increment in intake of fruit, vegetables, and fruit and vegetables combined [65]. The association seems to be non-linear, with stronger reductions in risk at lower levels of intake. This effect was significant for both ischemic and hemorrhagic stroke [65]. Persons with higher fruit and vegetable intake were more likely to be non-smokers, engaged in more physical activity, more highly educated, and were more likely to live in urban areas [66]. Other factors such as the use of pesticides and herbicides, water contamination, preferences methods of cultivation, types of fruits and vegetables commonly consumed, cooking methods, and affordability may play a role in primary prevention.
In a meta-analysis whole grain intake was associated with a reduction in cardiovascular disease, coronary heart disease, and all-cause mortality, but not with stroke [67].
Fish, Omega 3 Fatty Acids
The consumption of oily fish or long chain omega 3 fatty acids has been suggested to decrease the risk of vascular disease by lowering serum lipids, decreasing BP, decreasing platelet aggregation, improving vascular reactivity, and decreasing inflammation. Ecological studies raised the concern that high fish consumption may increase the risk of hemorrhagic stroke. A meta-analysis of 21 prospective cohort studies (675 048 participants, 25 320 cerebrovascular events) found a moderate but significant reduced risk of stroke when comparing the lowest category of fish intake with the highest (RR 0.88; 95% CI 0.84–0.93) [68]. The effects for ischemic and hemorrhagic strokes were similar. Long chain omega 3 supplements had, however, no influence on stroke incidence in RCTs which were mainly secondary prevention trials [64, 68]. Benefits of fish consumption compared to omega 3 fatty acids supplements may come from additional nutrients of fish (e.g. vitamins B and D) or a reduced intake of unhealthy food (red meat), or result from an association with higher socioeconomic status. Concerns were raised on the negative effect of methyl mercury contamination in fish. An evaluation of all risks and benefits of fish intake indicates that for modest fish consumption (one to two servings/week) the benefits of fish intake exceed the potential risks; people with very high consumption should limit some fish species with high mercury levels [69].
Sodium, Potassium, Calcium, and Magnesium
Observational studies found an association between high sodium intake and stroke incidence [70]. On the other hand, a meta-analysis of RCTs found only a weak evidence of salt reduction on cardiovascular events. This probably results from insufficient statistical power [71]. The observed positive effect of salt reduction is at least partly mediated by the well-studied positive relationship between salt intake and BP. A reduction in salt intake from an average high usual sodium intake level (201 mmol/day) to an average level (66 mmol/day) which is below the recommended upper level of 100 mmol/day (5.8 g salt) reduced systolic BP by 5.5 mmHg and diastolic BP by 2.9 mmHg in white hypertensive persons. In white normotensive individuals systolic and diastolic BP were reduced by 1.0 mmHg and 0.0 mmHg respectively [72]. A small number of studies suggest that beneficial effects are larger in Black and Asian persons. The Dietary Approaches to Stop Hypertension (DASH) trial, an RCT including 412 participants, found strong evidence for the benefit of low sodium intake [73]. Participants were randomized to one of three sodium intake levels and either the DASH diet (rich in vegetables and fruits, and low in dairy fat products and total and saturated fat and cholesterol) or a control diet (a typical American diet). Sodium reduction as well as the DASH diet reduced BP significantly.
Epidemiological studies found an inverse relationship between intake of potassium and risk of stroke, especially in ischemic stroke. In a meta-analysis of 16 prospective studies the lowest category of potassium intake compared to the highest category was associated with a 13% reduced risk of stroke when adjusted for BP [74]. A linear dose-response association was found up to a potassium intake of 90 mmol (≈3 500 mg)/day which was associated with the lowest risk of stroke. In particular, increased potassium intake reduces BP in people with hypertension without having an adverse effect on blood lipid concentration, catecholamine concentrations, or renal function [75].
Calcium intake was found to be inversely associated with BP; however, the association with stroke incidence is inconsistent between studies and meta-analyses. A recent meta-analysis of cohort studies found that high calcium and non-dairy calcium intake were not significantly associated with stroke risk, whereas high dairy calcium intake was associated with a 24% reduction in the risk of stroke [76]. The effect of calcium intake may differ between populations and may be confounded by other nutrients in dairy food.
Dietary magnesium has been found to be modestly inversely associated with the incidence of ischemic stroke in prospective cohort studies (RR for an increase in intake of 100mg magnesium/day 0.91; 95% CI 0.87–0.96) [77].
Coffee, Tea, and Chocolate
Coffee, tea, and chocolate consumption have been associated with lower rates of stroke. Favorable effects are probably not mediated by an effect on BP, but by the anti-oxidant capacity of polyphenolic compounds. A meta-analysis of 17 prospective cohort studies suggests a weak inverse non-linear association of coffee consumption with stroke with the strongest effect for 3.5 cups/day (RR 0.80; 95% CI 0.75-0.86) compared to 0 cups per day [78]. Nevertheless, there might be concerns on deleterious physiological effects in the hour after consumption that may trigger stroke onset. The consumption of green or black tea reduced the risk of stroke by 18% for each increment of 3 cups per day as shown by a meta-analysis of eight observational studies [79]. A meta-analysis of eight observational studies showed a 16% decreased relative risk of stroke in the highest compared to the lowest category of chocolate consumption and suggested a non-linear association with no more risk reduction when consuming above three servings (90g) per week [80]. However, no information of the type of chocolate consumed is available, and harmful effects resulting from the high sugar, saturated fat, and caloric content of commercially available chocolate should be considered.
Diet
Different nutrients and aliments cannot be seen independently of each other and thus the effect of different diets has been investigated. The DASH diet (see above) was associated with a significant decrease in BP [73]. A meta-analysis of three cohort studies found that adherence to a DASH-style diet significantly decreases stroke risk [81]. A Mediterranean-style diet rich in α-linolenic acid, olive oil, canola oil, fish, fruits, vegetables, and whole grains and low in saturated fat has been found to be successful for the prevention of cardiovascular diseases. In a meta-analysis of six observational studies adherence to a Mediterranean diet was associated with a significantly reduced incidence of stroke (RR highest versus the lowest 0.73; 95% CI 0.59–0.91) [82]. A meta-analysis based on three RCTs reported a protective effect of a Mediterranean diet compared to a control diet (RR 0.65; 95% CI 0.48–0.88) [83]. However, the number of stroke events was low (167) and was predominately contributed by the PREDIMED trial. The PREDIMED trial, a primary prevention trial in 7 447 participants at high cardiovascular risk, found that a Mediterranean diet supplemented with extra-virgin olive oil or mixed nuts can reduce stroke incidence when compared to a control group advised to reduce dietary fat [84]. On the other hand, an RCT including 48 835 women with dietary interventions consisting of total fat reduction to 20% of energy intake, and an increased intake of fruits, vegetables, and grain, did not result in a reduced incidence of coronary events and stroke [85]. This may suggest that the amount of total fats consumed may be less important than the type of fats.
A diet low in sodium, high in potassium, and rich in fruits and vegetables, whole grains, cereal fiber, and fatty fish has the highest potential to reduce stroke risk. Furthermore, coffee, tea, and chocolate consumption have been associated with lower rates of stroke.
Post-Menopausal Estrogen Replacement Therapy
Until menopause women generally suffer from a lower rate of vascular diseases, including ischemic stroke [86]. This has been attributed to a protective effect of estrogen and thus research has focused on the beneficial effect of post-menopausal hormone therapy for the prevention of cardiovascular diseases and stroke. However, the Women’s Health Initiative (WHI), a large RCT of 16 608 generally healthy post-menopausal women, showed that oral conjugated equine estrogen plus progestin increased the risk of ischemic stroke by 44% [87]. Recent meta-analyses confirmed that estrogen alone or estrogen–progestin therapy had no effect for primary prevention of cardiovascular disease and increased stroke incidence [88, 89]. Recent but limited evidence suggests that the risk of stroke is not increased in healthy post-menopausal women under the age of 60 at low risk of cardiovascular disease taking a low dose of transdermal estradiol (≤50 µg/day) for a short time. Thus, current research focuses on optimal dose, duration, route of administration, and timing of initiation of hormone replacement therapy [86, 88, 89].
Selective estrogen receptor modulators (SERMs) are a new class of drugs used for hormone replacement therapy lacking the steroid structure of estrogens, but able to bind directly to estrogen receptors. To date, few studies have investigated the effect of SERMs on stroke risk. In an individual participants’ meta-analysis of nine trials comparing SERMS with placebo for the prevention of breast cancer no effect was found on stroke risk, but the risk of venous thromboembolic events was significantly increased for all SERMS [90].
In the Raloxifene Use for The Heart (RUTH) trial including 10 101 post-menopausal women with coronary heart disease or multiple risk factors for coronary heart disease, the risk of fatal stroke was increased [91]. A meta-analysis (nine trials) investigating the risk of ischemic stroke in tamoxifen treatment for breast cancer found an increase of overall (RR 1.4; 95% CI 1.1–1.7) and ischemic stroke risk (RR 1.8; 95% CI 1.4–2.4) [92]. In the Post-menopausal Evaluation and Risk-reduction with Lasofoxifene (PEARL) trial, lasofoxifene reduced the risk of stroke, whereas the risk of deep vein thrombosis was increased [93].
Oral conjugated equine estrogen plus progestin increased the risk of ischemic stroke; so did raloxifene. The risk of stroke may not be increased with a low dose of transdermal estradiol, selective estrogen receptor modulators (SERMs), and lasofoxifene. Current research focuses on optimal dose, duration, route of administration, and timing of initiation of hormone replacement therapy.
Diseases and Pathological Conditions
Hypertension
Elevated BP is the best-documented and leading modifiable risk factor for stroke (Table 7.1). In the INTERSTROKE study the population-attributable risk for hypertension defined as self-reported hypertension or BP ≥140/90 mmHg was 46% for ischemic stroke and 56% for intracerebral hemorrhage [1]. Similarly, in the GBD 2013 64% of stroke burden (measured in DALYs) is attributed to high systolic BP [2]. High BP (BP >115/75 mmHg) is strongly and directly related to vascular and overall mortality without evidence of any threshold (Figure 7.4) [94]. Starting at a BP of 115/75 mmHg, stroke mortality risk increases steeply in an approximately log-linear relationship with BP [94]. Age attenuates this relationship and stroke risk increases with every 10 mmHg of systolic BP by 40–50%, 30–40%, and 20–30% for the age groups <60, 60–69, and ≥70 respectively [95].
Figure 7.4 Linear association between age-specific stroke mortality and usual blood pressure, data from a meta-analysis of prospective cohort studies.
Lowering BP substantially reduces the risk of stroke and major cardiovascular diseases. A recent meta-analysis of 123 RCTs (54 with stroke as outcome) showed RR reductions to be proportional to the magnitude of BP reduction: a reduction of 10 mmHg systolic BP reduced the relative risk of stroke by 26% [96] (Figure 7.5). The benefits of BP lowering were consistent and comparable across diverse BP baseline levels and between individuals with and without established cardiovascular disease.
Figure 7.5 The risk reduction in stroke events was found to be proportional to the magnitude of the systolic blood pressure reduction achieved (difference between study treatment groups) in a meta-analysis of randomized trials.
As a consequence, guidelines recommend lowering BP to 140/85 mmHg or below. The optimal BP target remains a point of controversy. Several trials including Systolic Blood Pressure Intervention Trial (SPRINT), Secondary Prevention of Small Subcortical Strokes (SPS3), and Action to Control Cardiovascular Risk in Diabetes (ACCORD) have tested whether more intensive BP targets (systolic BP <120 or <130 mmHg) compared to standard treatment (<140 mmHg) reduce the risk of major cardiovascular outcomes. Meta-analyses comparing intensive to less intensive targets found significant reductions of stroke events ranging between 18% and 35% [97]. Controversially, the Heart Outcomes Prevention Evaluation (HOPE)-3 trial testing in a 2×2 factorial design the effect of BP and cholesterol-lowering medication on cardiovascular outcomes in individuals at intermediate risk without cardiovascular disease found an effect of BP lowering on stroke events only for the subgroup of participants in the highest BP tertile group (baseline systolic BP >143.5 mmHG) [98]. Thus, the current ACC/AHA high BP clinical practices guidelines recommend more intensive BP targets (<130/80 mmHg) for primary prevention of cardiovascular diseases only in persons at high cardiovascular risk, i.e. in adults with confirmed hypertension (BP≥ 140/90 mmHg) or with a 10-year atherosclerotic cardiovascular disease risk of 10% or higher [99]. Lifestyle changes are recommended as part of the therapy. A combination of two or more anti-hypertensive agents is often necessary and preferable to achieve these targets [99]. In a recent network meta-analysis, no first-line anti-hypertensive treatment (angiotensin-converting enzyme inhibitors [ACEIs], calcium antagonists, angiotensin-receptor blockers [ARBs], calcium channel blockers [CCBs], or beta-blockers) was superior to thiazide or thiazide-like diuretics (THZs) for stroke prevention [97]. The RRs of stroke were 1.1 (95% CI 0.98–1.4) for ACEIs; 1.1 (95% CI 0.88–1.4) for ARBs; 1.3 (95% CI 1.1–1.6) for beta-blockers; and 0.96 (95% CI 0.83–1.2) for CCBs, compared with THZs.
As the strength in the association between BP and stroke risk attenuates with age, one might expect differences in the effect of BP-lowering drugs in older patients. Additionally, the prevalence of systolic hypertension (systolic BP >140 mmHg and diastolic BP <90 mmHg) increases with age. In elderly subjects, controlling hypertension regardless of whether or not it is isolated systolic hypertension has been shown to be beneficial [100]. A meta-analysis including nine trials comprising persons aged 60 years or older provided strong evidence that BP control to less than 150/90 mmHg reduces stroke (RR 0.74; 95% CI 0.65–0.84), and low- to moderate-strength evidence for lower BP targets (≤140/85 mmHg; RR for stroke 0.79; CI 0.59–0.99) [101]. The Hypertension in the Very Elderly Trial (HYVET), a randomized controlled trial, showed that even hypertensive patients older than 80 years benefit from BP-lowering therapy by a reduction in non-fatal stroke rate (RR 0.7; 95% CI 0.5–1.0) and stroke mortality (RR 0.6; 95% CI 0.4–1.0) [102].
Elevated blood pressure (BP) is the best-documented treatable risk factor for stroke. Lowering BP reduces stroke risk by 26% for every 10 mmHg systolic BP reduction.
Diabetes Mellitus
Diabetes is a well-documented risk factor for stroke. In a meta-analysis of 102 prospective studies (530 083 participants), the hazard ratio for ischemic stroke was 2.3 (95% CI 2.0–2.7) and 1.6 (95% CI 1.2–2.1) for hemorrhagic stroke in people with versus those without diabetes [103].
Nevertheless, there is no evidence from RCTs that anti-diabetic drugs can reduce stroke incidence or recurrence. Furthermore, a meta-analysis of nine randomized controlled trials (59 197 participants, 2 037 stroke events) suggests that intensive glycemic control compared to standard care does not appear to reduce stroke incidence [104]. However, evidence is insufficient because studies on conventional anti-diabetic agents focused mainly on glucose-lowering properties and information on cardiovascular outcome is limited. A meta-analysis investigating the effect of metformin, the first-choice drug for the treatment of type 2 diabetes, found no effect on stroke (RR 1.04; 95% CI 0.73–1.48), but was based on only 111 stroke events [105]. Recent trials suggest that newer glucose-lowering agents such as the glucagon-like peptide-1 (GLP-1) semaglutide may be promising for the reduction of stroke risk [106].
In addition to an increased stroke risk, subjects with type 2 diabetes have an increased prevalence of other stroke risk factors such as obesity, hyperlipidemia, hypertension, obstructive sleep apnea, and atrial fibrillation. Intensive multiple risk factor therapy and especially lifestyle modification can decrease the risk of cardiovascular events (including stroke) in people with type 2 diabetes [107]. However, the largest study, the Look AHEAD (Action for Health in Diabetes) trial, which randomized 5 145 patients with type 2 diabetes to an intensive lifestyle intervention or to usual care found no effect on cardiovascular morbidity and mortality after a median follow-up of 9.6 years [108].
Hypertension and diabetes are highly correlated: approximately 80% of adults with diabetes are hypertensive. Because of limited evidence, the optimal BP treatment goal in diabetic patients is still debated. Assuming an atherosclerotic cardiovascular disease risk ≥10% in hypertensive diabetic patients, BP should be lowered to less than 130/80 mmHg according to guidelines [99].
Dyslipidemia in type 2 diabetes is characterized by an increased blood triglyceride concentration and reduced HDL cholesterol concentration. However, total and low-density lipoprotein (LDL) cholesterol concentrations do not differ from the general population [109]. Treatment with statins reduces LDL cholesterol and stroke risk similarly to results seen in non-diabetic persons. A meta-analysis of four randomized controlled trials (10 187 participants) testing statin therapy for primary prevention of major cardiovascular and cerebrovascular events in diabetic patients found a significant relative risk reduction in fatal and non-fatal stroke (RR 0.69; 95% CI 0.51–0.92) [110].
There is insufficient evidence that improving glucose control reduces stroke.
Blood pressure (BP) target in hypertensive diabetic patients should be less than 130 mmHg systolic and 80 mmHg diastolic BP.
Treatment of diabetic patients with statins reduced the risk of stroke by 31%.
Dyslipidemia
A meta-analysis of older epidemiological studies found no relationship between total serum cholesterol level and overall stroke incidence [111]. This might be due to different relationships for ischemic and intracerebral hemorrhages. In many but not all prospective observational studies stroke risk was found to be positively associated with total and LDL serum cholesterol level in ischemic stroke, but negatively in intra-cerebral hemorrhages [112, 113]. Several prospective cohort studies suggest that the incidence of ischemic stroke is negatively associated with HDL cholesterol levels and positively with triglycerides [112]. However, the evidence is mixed. No such relationship was found for hemorrhagic stroke [113]. The relationship between ischemic stroke and cholesterol might further be obscured by different associations in different stroke subtypes, with the strongest positive relationship found in large atherosclerotic strokes [112].
Furthermore, age, sex, and vascular risk factors can modify the relationship between blood cholesterol and stroke mortality and morbidity [111, 114]. A meta-analysis of 61 observational prospective studies analyzed the influence of blood cholesterol on vascular mortality by distinguishing different age classes, sex, and different levels of BP [114]. A weak positive association between total blood cholesterol and ischemic and total stroke mortality was only found in the middle age group (40–59 years) and may be accounted for by an association between total cholesterol and BP. For systolic blood pressure levels below 145 mmHg, the association between cholesterol and stroke mortality was positive; for higher BP levels the relationship was negative.
In contrast to the weak and partly inconsistent findings from epidemiological studies, randomized controlled trials found a clear positive effect of cholesterol-lowering statin (3-hydroxy-3-methylglutaryl coenzyme A [HMG-CoA] reductase inhibitors) therapy on the incidence of major vascular events, including ischemic stroke (Figure 7.6 [115]). This effect may not only derive from lower lipid levels, but also from anti-inflammatory and anti-thrombotic properties of statins. In a recent meta-analysis of 13 RCTs, including the recent large HOPE-3 trial, in adults without prior cardiovascular events statin treatment reduced the risk of fatal and non-fatal stroke by RR 0.71 (95% CI 0.62–0.82) when compared to placebo, with no evidence for serious harm [116]. In particular, this analysis showed that all subgroups based on sex, age, baseline lipid level presence of hypertension, metabolic syndrome, or diabetes had similar benefits [116]. Observational studies have raised concerns about a higher incidence of hemorrhagic strokes. A meta-analysis of 31 randomized trials found no difference in incidence of intracerebral hemorrhages (676 cases) in the active treatment group compared to controls (OR 1.08; 95% CI 0.88–1.32; p = 0.47) [117]. The risk of intracerebral hemorrhages was not related to the degree of LDL reduction. However, the risk of bleedings might be different in Asian populations, where hemorrhagic strokes are more common. Recent studies suggest that myalgias may occur in a substantial number of patients treated with statins. In contrast, severe myopathy is a rare and generally self-limited side-effect of statin medications – especially in primary prevention trials using a low to moderate dose of statins [118].
Figure 7.6 Reductions in major vascular event were proportional to absolute LDL cholesterol reductions in a meta-analysis of randomized trials of routine statin therapy versus no routine statin use and of more intensive versus less intensive regimens.
Because of the linear relationship between risk reduction in stroke and LDL cholesterol level, a more intensive statin therapy might be indicated. A meta-analysis of 10 mainly secondary prevention trials for coronary heart disease (41 778 participants) found a significant reduction of fatal and non-fatal stroke (RR 0.86; 95% CI 0.77–0.96) for more intensive statin therapy compared to less intensive therapy [119].
Among non-statin interventions used to lower LDL cholesterol levels are fibrates, niacin, cholesteryl ester transfer protein (CETP) inhibitors, proprotein convertase subtilisin/kexin type 9 (PCSK9), and lifestyle modifications. Fibrates are effective in elevating HDL cholesterol, lowering triglyceride concentration, and reducing LDL cholesterol. A meta-analysis of six primary prevention trials including 16 135 individuals without established cardiovascular disease that compared fibrate therapy with placebo or usual care found a small but statistical significant reduction for the combined vascular outcomes including stroke (RR 0.84, 95% CI 0.74–0.96) [120].
The results of another Cochrane review of 23 RCTs suggests that niacin does not reduce cardiovascular mortality or cardiovascular events, nor the number of fatal or non-fatal strokes (RR 0.95, 95% CI 0.74–1.22; 33 661 participants, seven studies), but is associated with side-effects [121].
New lipid-lowering drugs such as the PCSK9 inhibitors alirocumab, bociziumab, evolocumab, or the CETP-inhibitor anacetrapib have shown to achieve small but significant absolute risk reductions of predominantly cardiovascular outcome events in high-risk patients already receiving statin therapy [122, 123]. A meta-analysis of 35 RCTs comprising 45 539 patients found that compared with no PCSK9 inhibitor therapy, treatment with a PCSK9 inhibitor was associated with a lower rate of stroke (OR 0.80; 95% CI 0.67–0.96) [122]. Long-term confirmatory trials are needed, particularly in PCSK9 inhibitors.
Lifestyle modifications including weight loss, physical activity, and diet are recommended to improve lipid profile.
Randomized controlled trials found a clear positive effect of cholesterol-lowering statin therapy on the incidence of ischemic stroke. Niacin or fibrates showed no effect on stroke incidence. Treatment with a PCSK9 inhibitor was associated with a lower rate of stroke, but long-term confirmatory trials are needed.
Related posts:
Chapter 4 – Imaging for Prediction of Functional Outcome and for Assessment of Recovery
Chapter 8 – Cardiac Diseases Relevant to Stroke
Chapter 16 – Stroke and Dementia
Chapter 19 – Acute Therapies for Stroke
Chapter 22 – Management of Acute Ischemic Stroke and its Late Complications
Chapter 21 – Intensive Care of Stroke
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
