Lifestyle Management Evidence Tables and Reference List
A healthy lifestyle, which includes a healthy balanced diet, exercise, weight control, reduction and avoidance of alcohol and tobacco, reduces the risk of an initial stroke and the risk of a subsequent stroke for patients with a prior history of stroke. Data from the Global Burden of Disease Study 2013 (Murray et al, 2013) were used to estimate the population-attributable fraction (PAF) of stroke-related disability-adjusted life-years (DALYs) associated with 17 potentially modifiable risk factors. While global estimates were provided, data from separate countries were also reported. Stroke burden among Canadians was attributed to a variety of modifiable risk factors, including 20% for diets low in fruits and vegetables, 13% for diets high in sodium, 11% for low levels of physical activity, 28% for a body mass index greater than 23.0, and 13% for tobacco use. These results are consistent with other reports. The recent INTERSTROKE 2 study (O'Donnell et al 2016) reported that among 10 risk factors, the odds of all stroke were 2.5 times higher among persons with a self-reported history of hypertension, 2 times higher among heavy alcohol consumers and over 1.5 times higher for tobacco smokers. The associated population attributable risk estimates were 34%, 5.8% and 12%, respectively.
Examining the relationship between stroke risk and diet is challenging, in large part due to the limitations in methods for collecting long-term dietary intake and controlling for potential confounders. The results of studies evaluating individual dietary components (fruit/vegetable consumption, fats, dairy products and whole grains) and dietary patterns of eating have yielded ambiguous results.
There is evidence to suggest that regular consumption of fruits and vegetables reduces the risk of stroke. The results from the China Kadoorie Biobank Study (Du et al. 2016) included 512,891 adults, aged 35-74 years, without a history of cardiovascular disease or treatment for hypertension. During 3.2 million person-years of follow-up, the incidences of both ischemic and hemorrhagic stroke were significantly lower among those who consumed fruit at least monthly. The reduction was dose-dependent, such that daily consumption was associated with the lowest risk for both stroke types. The Hazard Ratios for daily consumption were 0.75 (ischemic stroke) and 0.64 (hemorrhagic stroke). Data from the Global Burden of Disease Study 2013 (Feigin et al. 2016) was used to estimate the population-attributable fraction (PAF) of stroke-related disability-adjusted life-years (DALYs) associated with 17 potentially modifiable risk factors, including diets low in fruits and vegetables. While data from 188 countries was reported, country-specific estimates were also provided. In Canada, 20.4% (95% uncertainty interval 9.7%-31.5%) of the stroke burden was attributed to diets low in fruits, while 19.5% (95% uncertainty interval 14.4%-25.5%) was attributed to diets low in vegetables. In a cohort study including 175,000 participants, Sharma et al. (2013) reported no protective effect of consuming fruit or vegetables in either men or women during 8 years of follow-up. The mean number of servings of fruit and vegetables between those who died of stroke was similar to all others, assessed by determining whether participants were compliant with the USDA’s food pyramid. In a case-control study, O’Donnell et al. (2010) reported that increased consumption of fruit was associated with a decreased risk of stroke (adj OR of tertile 1 vs. 3=0.61, 99% CI 0.66-0.91), while increased consumption of vegetables was not (adj OR of T3 vs. T1=0.91, 99% CI 0.75-1.00). A meta-analysis (He et al. 2006) that included 20 studies and 760,629 participants, with follow-up ranging from 4-37 years, reported the risk of stroke was significantly lower in the groups associated with the highest intake of fruits and vegetables (Total combined fruit and veg: RR=0.79, 95% CI 0.75-0.84; Fruit: RR=0.77, 95% CI 0.71-0.84; Vegetables: RR=0.86, 95% CI 0.79-0.93). For every increase of 200 g/day of vegetables, stroke risk was decreased by 11% (RR=0.89, 95% CI 0.81-0.98). The corresponding decrease in stroke risk for fruit was 32%.
The role of dietary fat as a risk factor for stroke remains unclear. Siri-Tarino et al. (2010) conducted a systematic review & meta-analysis that included the results of 21 prospective cohort studies, of which stroke was the outcome in 8 studies (n=179,436). The mean follow-up periods ranged from 8-23 years. There was no increased risk of stroke associated with the highest intakes of saturated fat compared with the lowest (adjusted RR=0.81, 95% CI 0.62-1.05, p=0.11). Results from the Prospective cohort study Multi-Ethnic Study of Atherosclerosis (MESA), suggest that the source of saturated fat is a greater indicator of cardiovascular risk (De Oliveira Otto et al. 2012). While stroke was not an outcome assessed in this study, saturated fat from dairy sources was found to be protective for incident cardiovascular disease, while the risk was increased for consumption of saturated fat from meat sources. He et al. (2003) did not find significant associations between amount of total fat, source of fat (animal or vegetable), type of fat (saturated, unsaturated, monounsaturated, polyunsaturated, trans fat or cholesterol) or selected high-fat foods, including red meat, high-fat dairy products, nuts and eggs, and incidence of ischemic or hemorrhagic stroke. This prospective cohort study included data from 43,732 men aged 40-75 years from the Health Professionals’ follow-up study who were free of cardiovascular disease and diabetes at baseline. The consumption of trans fat is generally thought to be associated with negative health outcomes. Using data from the REasons for Geographic And Racial Differences in Stroke (REGARDS) study, including 17,107 participants ≥45 years of age and without a history of stroke, Kiage et al. (2014) examined the relationship between incident stroke and trans-fat consumption. During a median of 7 years of follow-up, the risk of ischemic stroke was reported to be elevated significantly for men (HR=1.13, 95% CI 1.00-1.28), but not for women (HR=0.93, 95% CI 0.77-1.12).
The results from two recent systematic reviews (Hu et al. 2014, Qin et al. 2015), including the results from 15 and 22 cohort studies, respectively, suggest that dairy consumption may be protective for stroke (RR=0.80, 95% CI 0.76-0.84 and RR=0.87, 95% CI 0.77-0.99). While increased consumption of dairy products in general was associated with a lower risk of stroke, in sub group analyses the effect was most pronounced for low-fat dairy, cheese and fermented milk products. Hu et al. (2014) reported a non-linear dose-response relationship between milk consumption and stroke risk whereby 200 mL/day was most protective (RR=0.82, 95% CI 0.79-0.86). Larsson et al. (2012) also reported that consumption of low-fat dairy products was associated with a decreased risk of all stroke (RR=0.88, 95% CI 0.80-0.97) and ischemic stroke (RR=0.87, 95% CI 0.78-0.98) in a cohort study including 74,961 Swedish men and women, aged 45-83 years without a history of stroke.
In terms of reductions in stroke risk associated with different dietary patterns, Agnoli et al. (2011) compared adherence to four commonly-recognized diet regimes and their impact on stroke risk, including the Healthy Eating Index 2005 (HEI-2005), Dietary Approaches to Stop Hypertension (DASH), Greek Mediterranean Index, and the Italian Mediterranean Index. There was an inverse relationship between adherence to each of the regimens and stroke occurrence. Overall, the Italian Mediterranean Index was the most protective (HR = 0.37, 95% CI = 0.19–0.70). A systematic review & meta-analysis was conducted by Psaltopoulou et al. (2013) to examine the protective effects associated with adherence to the Mediterranean diet. Of the 11 studies that were included that assessed stroke as an outcome, high adherence to a Mediterranean diet was associated with reduced risk of total stroke and ischemic stroke (total stroke: RR=0.71, 95% CI 0.57-0.89; ischemic stroke: RR=0.52, 95% CI 0.28-0.96). One of the key components of the Mediterranean diet is olive oil, which has been shown to decrease the risk of cardiovascular diseases. The Prevención con Dieta Mediterránea Trial (PREDIMED) evaluated the benefits of 2 types of Mediterranean diet, increased consumption of extra-virgin olive oil or mixed nuts, as compared to a control group in which participants were advised to follow a low-fat diet (Estruch et al. 2013). After a median follow-up of 4.8 years, the two Mediterranean diets were associated with 30% reductions in the primary outcome, a composite of myocardial infarction, stroke, or death from cardiovascular causes. The majority of this protective effect was driven by a reduction in stroke events. The results of the PREDIMED study were included in a systematic review (Martınez-Gonzalez et al. 2014) specifically examining the protective effect of olive oil. For each 25 g/day increase in olive oil consumption there was a significant reduction in the risk of stroke (RR=0.76, 95% CI 0.67-0.86, p<0.001). A systematic review and meta-analysis authored by Soedamah-Muthu et al. (2013) provides evidence of the reduced risk of stroke associated with a dietary approach similar to the Mediterranean-style diet, the DASH style diet, which is characterised by fruits, vegetables, and low-fat or non-fat dairy, as well as less-refined grains. High adherence was protective for the development of cardiovascular disease (RR=0.80, 95% CI 0.74-0.86). Larsson et al. (2016) also reported that high adherence to a modified DASH diet was associated with a reduced risk of ischemic stroke, particularly among women. The study included a population-based sample of almost 75,000 individuals without history of stroke, heart disease or cancer, who were followed for an average of 11.9 years.
Vitamin B supplementation to Reduce the Risk of Recurrent Stroke
Hyperhomocysteinemia has been associated with premature atherosclerosis and an increased risk of cardiovascular events, including stroke. Since low serum levels of B vitamins, including B6, B12 and folic acid are associated with elevated homocysteine levels, supplementation has been examined as a potential means to reduce stroke risk. Two large trials have been published which included persons exclusively with recent stroke. In the VITAmins TO Prevent Stroke (VITATOPS) trial (Hankey et al. 2010) participants were randomized to receive B vitamins (2 mg folic acid, 25 mg vitamin B6, and 0.5 mg vitamin B12) or placebo for the duration of the trial. After a median duration of follow-up of 3.4 years, there was a borderline significant reduction in the risk of the composite outcome, which included stroke, MI and vascular death in the vitamin B group (15% vs. 17%, RR=0.91, 95% CI 0.82-1.00, p=0.05, absolute risk reduction of 1.56%, 95% CI -0.01-3.16). The risk of fatal or nonfatal stroke was not reduced significantly with vitamin supplementation (9% vs. 10%, RR=0.92, 95% CI 0.81-1.06, p=0.25). Toole et al. (2004) randomized 3,680 patients with a total homocysteine level ≥25th percentile to receive high-dose B vitamins (25 mg B6, 0.4 mg B12, and 2.5 mg folic acid) or low-dose B vitamins (200 µg B6, 6 µg B12, and 20 µg folic acid) in the Vitamin Intervention for Stroke Prevention (VISP) trial. The mean duration of follow-up was 20.3 months. There was no significant difference between groups in the relative 2-year risk of recurrent cerebral infarction (8.1% vs. 8.4%, RR=1.0, 95% CI 0.8-1.3).
It is well documented that a consistently high dietary sodium intake is associated with elevated blood pressure, while modest decreases may lower blood pressure and reduce stroke risk. Mozaffarian et al. (2014) used various data sources and national-level surveys to estimate that, in 2010, 99% of all adults in the world exceeded the WHO recommendations of 2.0 g/day. Worldwide, the mean global level of sodium intake was 3.95 g/day. An estimated 1.65 million deaths were attributed to sodium intake above the recommended level, of which 685K (42%) were caused by stroke. Feigin et al. (2016) estimated that 22.6% of the global stroke burden was attributed to diets high in sodium (12.6% in Canada). In a Cochrane review, He et al. (2013) examined 34 RCTs (n=3,230) comparing the effect of moderately restricted sodium intake (2.3-7.0 g/day or 40-120 mmol/day urinary sodium excretion) for a minimum of 4 weeks with usual intake over the same duration. The mean difference in sodium intake between groups was 1,955 mg per day, which was associated with a significant decrease in SBP (-4.18 mmHg, 95% CI -5.18 to -3.18; p<0.001) and DBP (-2.06 mmHg, 95% CI -2.67 to -1.45; p<0.001). Results were similar in a subgroup analysis of 22 trials that included 990 patients with hypertension. Reduced intake was associated with a significant reduction in both SBP (-5.39 mmHg, 95% CI -4.15 to -6.62; p<0.001) and DBP (-2.82 mmHg, 95% CI -2.11 to -3.54; p<0.001). Abuerto et al. (2013) identified 36 RCTs (n=5,508; n with hypertension=1,478) also comparing the effects of decreased sodium vs. higher sodium intake. The mean between group difference in sodium intake was ≥40 mmol/day. Reduced sodium intake was associated with a mean SBP reduction of 3.39 mm Hg (95% CI 2.46 to 4.31) in all participants and a mean SBP reduction of 4.06 mm Hg (95% CI 2.96 to 5.15) in participants with hypertension. In trials where the relative sodium reduction of subjects in the intervention group was <1/3 of the control group, there was a significant reduction in both SBP (MD= -1.45, 95% CI -2.29 to -0.60) and DBP (MD= -0.74, 95% CI -1.28 to -0.19). In trials where the relative sodium reduction of subjects in the intervention group was ≥1/3 of the control group, the reductions in both SPB and DBP were even greater (SBP: MD= -3.79, 95% CI -4.82 to -2.75 and DBP: MD= -1.68, 95% CI -2.34 to -1.02). There is evidence of a U-shaped pattern associated with sodium intake and stroke risk/mortality. Graudal et al. (2014) included the results of 23 cohort studies (n=274,683) and reported that usual daily sodium intake (115 -215 mmol) was associated with a significantly lower risk of all-cause mortality compared with low-sodium intake (<115 mmol), with no effect on stroke risk. High sodium intake (>215 mmol) was associated with an increased risk of both stroke and all-cause mortality, compared with usual sodium intake. O’Donnell et al. 2014 reported that sodium intake between 3 g (130 mmol) and 6 g (260 mmol) per day was associated with a lower risk of death and cardiovascular events than either a higher or lower level of sodium intake.
Physical activity is an important modifiable lifestyle factor that can play a protective role in both primary and secondary prevention of stroke. Using data from 188 countries, obtained from the Global Burden of Disease Study, Feigin et al. (2016) reported that 7.7% of the global stroke burden was attributed to low physical activity. In Canada, the estimate was 10.9%. The results from several large cohort studies provide some estimates of the magnitude of the protective effect of physical activity. Armstrong et al. (2015) included 1.1 million women who were participants of the Million Women Study, which investigated how various reproductive and lifestyle factors affect women’s health. Women who engaged in strenuous physical activity 1-3x/week had a lower risk of both intracerebral hemorrhage and ischemic stroke compared with women who rarely or never engaged in such activity. The effect was U-shaped such that the risk of stroke was not reduced significantly for women who engaged in strenuous activity more than three times per week. In the REGARDS study, a large prospective cohort study including 30,239 US residents aged ≥45 years, McDonnell et al. (2013) reported that the risk of stroke was increased in persons who engaged in no physical activity, compared to persons who exercised ≥4x/week (HR= 1.20, 95% CI, 1.02–1.42). In phase 1 of the INTERSTROKE case-control study, O’Donnell et al. (2010) reported that regular physical activity was associated with a reduced risk of total and ischemic stroke (total stroke: OR=0.69, 99% CI 0.53-0.90, ischemic stroke: OR=0.68, 99% CI 0.51-0.91). In phase 2 of the INTERSTROKE study (O’Donnell et al. 2016, the pattern of results was similar. There was a decreased risk of total, ischemic and hemorrhagic stroke associated with regular physical activity. While Sattelmair et al. (2010) reported that increasing amount of time spent engaged in physical activity was not associated with decreased total stroke risk in 39,315 healthy women who had been participants of the Women’s Health Study (1992-1995), those who walked ≥2 hours per week had a 30% lower risk of any stroke than women who did not walk (RR=0.70 95% CI, 0.52 to 0.94). Additionally, women who reported walking at a brisk pace (4.8 km/hour) had a 37% lower risk (RR=0.63, 95% CI, 0.44 to 0.91) compared with women who did not walk. Lee et al. (2003) published a meta-analysis of 23 studies published between 1983 and 2002 examining the association between physical activity and stroke incidence or mortality and reported a dose-response relationship. Highly active individuals had a 27% lower risk of stroke than individuals who were designated as “low active.” Individuals who were designated as moderately active also had a significantly reduced risk of ischemic and hemorrhagic strokes when compared with low active individuals (RR = 0.80, 95% CI 0.69-0.91p < 0.001).
Evidence suggests there is an increased risk of stroke associated with being overweight or obese. Feigin et al. (2016) reported that 23.5% of the global stroke burden was attributed to high BMI (>23.0), while in Canada the estimate was 28.4%. Twig et al. (2016) included 2.3 million adolescents who were followed over time to examine the association between BMI and cardiovascular death. During 42,297,007 person-years of follow-up, there were 32,127 deaths, including 528 from stroke. Compared with the reference category (BMI percentile 5th-24th), the risk of death from stroke was significantly increased in the 3 highest BMI categories, in which the median BMI (men and women combined) were 24.4, 26.6 and 31.0, respectively (75th-85th: HR=1.42, 85th-94th: HR=1.81, ≥95th: HR=2.64). Saito et al. (2011) compared stroke risk in 32,847 men and 38,875 Japanese women, aged 45–74 years with no history of cardiovascular disease, who were of normal weight (BMI 23.0-24.9 kg/m2) with persons who had high BMIs (27.0 to 29.9 and ≥ 30.0). The risk of stroke significantly increased with increasing BMI (HR= 1.09 and 1.25 for men, and HR=1.29 and 2.16 for women, respectively, relative to healthy weight). In women, a weight increase of greater than 10% over the previous five years was also associated with increased stroke risk. Bazzano et al. (2010) reported similar findings in a study of 154,736 Chinese men and women ≥ 40 years. The risk for stroke increased significantly for persons considered overweight (BMI 25.0 to 29.9, HR=1.43, 95% CI 1.36-1.52) and for those who were obese (BMI≥ of 30, HR=1.72, 95% CI 1.55-1.91). In phases 1 and 2 of the INTERSTROKE case-control study, O’Donnell et al. (2010, 2016) reported that increasing weight-to-hip ratio was associated with increased risk of total stroke, ischemic stroke and hemorrhagic stroke. Hu et al. (2007) studied 49,996 men and women aged 25-74 years with no history of stroke or coronary heart disease. BMI, waist circumference and waist-hip measures were obtained at baseline and stroke risk was assessed after an average follow up of 9.5 years. The risk of all stroke and ischemic stroke were increased in both men and women with increasing BMI, while increased waist circumference and waist-to-hip ratio were risk factors for total and ischemic stroke in men, but not women.
Evidence from several studies suggest that light to moderate alcohol consumption may reduce the risk of stroke, while excessive consumption may increase risk. Zhang et al. (2014) used the results of 27 prospective studies including 1,425,513 adult participants to estimate this dose-response relationship. The relationship between ETOH dose and stroke risk was found to be j-shaped, with alcohol intake of 0-20 g/day associated with a significant reduction and intake above 40 g/day associated with increased risk. The association between alcohol consumption and risk of stroke may be different for men compared with women. Zheng et al. (2015) pooled the results from 23 cohort studies and found that, compared with the lowest or no alcohol groups, the risk of stroke was not significantly increased in men or women as alcohol consumption increased; rather, the risk of ischemic stroke was lower in men who were light drinkers and for women who were light or moderate consumers. In contrast, using the results from 26 studies, Patra et al. (2010) reported a dose-response relationship that was linear for hemorrhagic stroke, with increasing risk associated with increasing consumption, whereas there was a curvilinear relationship for ischemic stroke, with a protective effect of alcohol for low to moderate consumption and increased risk for higher exposure. Women who consumed 3 or more drinks on average/day had higher risk than men. O’Donnell et al. (2010) reported that moderate alcohol consumption (1-30 drinks/month) was associated with reduced risk of ischemic stroke (OR=0.79, 95% CI 0.63-1.00), but with an increased risk of hemorrhagic stroke (OR=1.52, 95% CI 1.07-2.16) compared with never/former drinkers. Binge drinking, or >30 drinks/month, was associated with an increased risk of ischemic and hemorrhagic stroke compared with never/former drinkers. In phase 2 of INTERSTROKE (O’Donnell et al. 2016) low or moderate ETOH intake was associated with significantly higher odds of total and hemorrhagic stroke compared with former/never drinkers, with no risk in the increase of ischemic stroke. A meta-analysis including 35 observational studies examining the effects of alcohol consumption on stroke risk over a follow-up period of 4-30 years revealed a J-shaped relationship between the amounts of alcohol consumed per day and the risk of ischemic stroke (Reynolds et al. 2003). Individuals who consumed <12 grams of alcohol per day had the lowest risk for ischemic stroke (RR = 0.80, 95% CI 0.67-0.96), while those having more than 60 grams/day had the highest risk (RR = 1.69, 95% CI 1.34-2.15) when compared with a group of abstainers.
Birth Control/Hormone Replacement Therapy
Women taking oral contraceptive or hormone replacement therapy (HRT) may be at an increased risk of stroke. Bath & Gray (2005) conducted a meta-analysis including the results from 28 RCTs and found that HRT was associated with significant increases in the risk of total stroke (OR =1.29, 95% CI 1.13 to 1.47), non-fatal stroke (OR=1.23, 1.06 to 1.44), stroke leading to death or disability (OR=1.56, 1.11 to 2.20), and ischaemic stroke (OR=1.29, 1.06 to 1.56). They also reported that hormone replacement therapy was not associated with hemorrhagic stroke (OR=1.07, 0.65 to 1.75) or transient ischaemic attack (OR=1.02, 0.78 to 1.34). Similarly, Renoux et al. (2010) reported that, compared to non-users, women using oral hormone replacement therapy within the previous year had a higher risk of stroke (RR= 1.28, 1.15-1.42). Use of oral HRT for >1 year was associated with increased risk of stroke (RR=1.35, 95% CI 1.20-1.52), but not for a duration of ≤1 year. High dose transdermal patch use was associated with an increased risk of stroke (RR=1.89, 95% CI 1.15-3.11), although low- dose patches were not (RR=0.95, 0.75-1.20).
In terms of elevated risk of stroke associated with hormonal forms of birth control, the evidence is equivocal. In a large cohort study including the results of over 1.6 million women between the ages of 15 and 49 years, Lidegaard et al. (2012) reported that current use of ethinyl estradiol at doses of 20 to 50 μg was associated with an increased risk of thrombotic stroke, compared with nonusers, while current use of progestin only was not. In a large cohort study of 49, 259 Swedish women aged 30-49, Yang et al. (2009) reported that the risk of fatal or nonfatal ischemic or hemorrhagic stroke was not significantly increased. The associations were not influenced by age at menarche nor with parity status.
Recreational Drug Use
The most commonly-used illicit drugs associated with increased stroke risk are cocaine, amphetamines, Ecstasy, heroin/opiates, phencyclidine (PCP), lysergic acid diethylamide (LSD), and cannabis/marijuana. These drugs may increase the risk for stroke through a variety of mechanisms, including hypertensive surges, vasospasm, enhanced platelet aggregation, vasculitis, accelerated atherosclerosis and cardioembolism. Kaku & Lowenstein (1990) reported that the risk of stroke associated with (any) drug abuse was significantly higher compared with non-drug users (RR=06.5, 95% CI 3.1-13.6). There was a strong temporal relationship whereby the risk was highest during the first 6 hours after use and decreased over time. Cheng et al. (2016) examined whether recent cocaine use increased the risk of stroke. Cocaine use within 24 hours of the reference date was associated with a significantly increased risk of ischemic stroke (OR=6.4, 95% CI 2.2-18.6, p<0.001), as was frequent use (≥1/week; OR=2.6, 95% CI 1.6-4.3, p<0.001). An increased risk of stroke associated with cocaine use was also reported by Westover et al. (2007) in a cohort of patients recently discharged from hospital. Previous cocaine use was associated with an increase in the risk of both hemorrhagic and ischemic stroke (OR=2.33, 95% CI 1.74-3.11and OR=2.03, 95% CI 1.48-2.79, respectively). In the same study, amphetamine use was also associated with an increase in the risk of hemorrhagic stroke (OR=4.95, 95% CI 3.24-7.55) and an increased risk of hemorrhagic stroke resulting in death (OR=2.63, 95% CI 1.07-6.50). The association between cannabis use and stroke does not appear to be as strong. While Barber et al. (2013) found no association between stroke and cannabis use (OR=1.59, 95% CI 0.71-3.70), Westover et al. (2007) reported that cannabis use was associated with an increased risk of ischemic stroke (OR=1.76, 95% CI 1.15-2.71) but not hemorrhagic stroke (OR=1.36, 95% CI 0.90-2.06), after adjusting for age, sex, ethnicity and current tobacco use.
Smoking is a major risk factor for cardiovascular disease, including stroke and heart attacks. Smokers are significantly more likely to have a stroke compared with non-smokers. It has been estimated that globally, 20.7% of the stroke burden is attributable to tobacco use (Feigin et al. 2016). There appears to be a dose-response relationship between increased cigarette smoking and stroke risk. In The Physician’s Health Study (Robbins et al. 1994), the risk of non-fatal stroke was significantly higher among those currently smoking ≥20 cigarettes/day compared with those who never smoked. (RR=2.52, 95% CI 1.75 to 3.61). For those currently smoking <20 cigarettes/ day, the stroke risk remained elevated (RR=2.02, 95% CI 1.23 to 3.31). A recent systematic review & meta-analysis (Peters et al. 2013) that reported sex-specific risk of current smokers vs. non-smokers included the results from 81 prospective cohort studies, which represented 3,980,359 persons. The prevalence of current smoking ranged from 8% to 59% in men and from 1% to 51% in women. Most studies reported higher smoking rates among men. Over the duration of follow up, which ranged from 6-40 years, there were 42,401 strokes. The risk of stroke was higher in current smokers compared with non-smokers in both women: (RR=1.83, 95% CI 1.58-2.12) and men (RR=1.67, 95% CI 1.49-1.88). The risk of stroke was also higher in former smokers compared with never smokers (women: RR=1.17, 95% CI 1.12-1.22; men: RR=1.08, 95% CI 1.03-1.13). The risk of hemorrhagic, but not ischemic stroke, was significantly increased in women who smoked compared with men who smoked (RR=1.17, 95% CI 1.02-1.34, p=0.02). An increased risk of all stroke (OR=2.09, 99% CI 1.75-2.51), ischemic stroke (OR=2.32, 99% CI 1.91-2.81) and hemorrhagic stroke (OR=1.45, 99% CI 1.07-1.96) was also associated with current smoking in Phase 1 of the case-control INTERSTROKE Study (O’Donnell et al. 2010). In phase 2 of the study (O’Donnell et al. 2016), which included a larger sample size (26,919), the risk of ischemic stroke was higher among current smokers compared with the risk of hemorrhagic stroke (OR=1.93, 99% CI 1.69-2.21 vs. OR=1.14, 99% CI 0.95-1.36). The risk of both stroke types increased with the number of cigarettes smoked daily. Results from the Cardiovascular Health Study (Kaplan et al. 2005) including persons over the age of 65 years, indicated that smoking was associated with a significantly increased risk for stroke recurrence (HR= 2.06; 95% CI, 1.39–3.56).
Both pharmacological agents and behavioural intervention strategies have proved effective as smoking cessation interventions. A Cochrane review of reviews that examined the effectiveness of pharmacological treatments to promote smoking cessation in adults was included the results of 12 Cochrane reviews, aggregating the results from 267 RCTs, 101,804 participants (Cahill et al. 2013). Treatments evaluated included nicotine replacement products, such as gums, transdermal patches, nasal sprays or inhalers, the non-tricyclic antidepressant, bupropion and varenicline, a nicotinic receptor partial agonists. Compared with placebo, all forms of therapies significantly increased the odds of sustained smoking cessation (odds ratios ranged from 1.82-2.88). Varenicline was superior to single forms of nicotine replacement therapy (OR= 1.57, 95% Credible interval [Cred I] 1.29 to 1.91) and was also superior to bupropion (OR= 1.59, 95% CredI 1.29 to 1.96). The odds of serious adverse events (chest pains and heart palpitations) associated with nicotine replacement therapy were significantly increased (OR= 1.88, 95% CI 1.37- 2.57). The most common side effects associated with bupropion were insomnia, occurring in 30% to 40% of patients, dry mouth (10%) and nausea. The main serious adverse event was seizures. The main adverse event for varenicline was mild-moderate nausea, which subsided over time and was rarely reported. Typical drop-out rates due to adverse events ranged from 7% to 12%.
Non-pharmacological and combination therapy have been shown to be effective in achieving sustained smoking cessation. A recent Cochrane Review authored by Stead & Lancaster (2012a) evaluated behavioral support with the addition of the availability of pharmacotherapy compared with a control condition receiving usual care or brief advice or less intensive behavioural support. The results from 41 RCTs including participants from both community and healthcare settings, most of whom smoked >20 cigarettes/day, were included. Most studies supplied nicotine replacement therapy (provided as patch or gum) while behavioural support was typically provided by specialists in cessation counselling, but was also provided by peer counsellors, trained nurses and usual care providers and took the forms of telephone, mail, individual and group sessions. Combination therapy was associated with the greatest chance of cessation of smoking at 6 months (RR=1.82, 95% CI 1.66-2.00, p< 0.0001). In studies that recruited participants from healthcare settings, the probability of success was greater (RR=2.06 vs. 1.53). There was no association between number of sessions provided and success of quitting (1-3 vs. 4-8 vs. >8) or the planned duration of contact (total minutes) (up to 30 vs. 31-90 vs. 91-300 vs. >300).
Mullen et al. (2016) examined the use of the Ottawa Model’ for Smoking Cessation (OMSC), a systematic approach to tobacco dependence treatment delivered within healthcare settings, which included in-hospital counselling, and pharmacotherapy follow-up support post hospitalization. At one and two years, the cumulative incidences of death and all-cause re-hospitalizations, and smoking-related readmissions were significantly lower in the OMSC group. All-cause emergency department visits were also significantly reduced in the intervention group. In this trial patients in the control group were randomized to usual care, which generally consisted of a self-help pamphlet.
Motivational interviewing, by itself has also shown to be an effective strategy to achieve sustained smoking cessation. Using the results from 14 RCTs, Lai et al (2010) examined the use of 1-4 sessions (15-45 minutes/session) of motivational interviewing (MI) compared with control groups who received brief advice, or routine care. Motivational interviewing was associated with a significantly increased probability of achieving long-term smoking cessation (RR 1.27, 95% CI 1.14- 1.42). Chances of success were greater when delivered by a general practitioner, compared with a nurse or counsellor. While both single compared and multiple sessions were both effective, sessions of >20 minutes duration were more effective compared with shorter sessions (RR= 1.31, 95% CI 1.16 to 1.49 vs. 1.14, 95% CI 0.80 to 1.16).
The use of electronic cigarettes (e-cigarettes) has increased in recent years, and remains controversial. They may be used as an alternative to conventional cigarettes or as an aid in smoking cessation programs. The current practice recommendations make no statements regarding their use. Although the use of e-cigarettes has been shown to significantly reduce the use of conventional cigarettes, in persons who wish to quit smoking (and in those with no desire to quit), data regarding their safety are limited.
Compliance with Secondary Prevention Measures
Since rates of recurrent stroke, and other vascular disorders are known to be significantly elevated during the first four years after hospitalization for first stroke (Feng et al. 2010), and potentially modifiable risk factors represent approximately 90% of the population-attributable risk for stroke (O’Donnell et al. 2016), secondary prevention measures represent an important opportunity to reduce the risk. While the effectiveness of many of the interventions designed to prevent recurrent stroke, including medications associated with hypertension, diabetes, dyslipidemia and cardiac conditions (described in other sections of the guidelines) are well-established, their protective effects are diminished by poor compliance. Poor- or non-compliance to recommended medications may be due to several factors including inadequate or marginal health literacy, number of co-morbid conditions, adverse effects of treatment and cost (MacLaughlin et al. 2012). Non-compliance with diet regimens and other lifestyle factors is often a result of the interplay between patients’ age, emotions, the reason they were given to control diet as well as their ability or desire to return for follow-up education (Travis 1997). Additional stroke-related factors, such as a lack of motivation, musculoskeletal issues, fatigue, and increasing age, may pose barriers to compliance with exercise programs or reduced leisure activities (Jurkiewicz et al. 2011). Therefore, early initiation of effective post stroke prevention strategies, maintained indefinitely with continuous monitoring by way of follow-up appointments, home visits or telephone check-in is essential. Bushnell et al. (2014) suggest a comprehensive model of stroke prevention, including the recognition of non-adherence, and understanding the factors associated with non-adherence. Moreover, clinician need to consider their patient demographic how they deliver secondary prevention treatment, with an emphasis on communication and education (Travis 1997, Hedegaard et al. 2015).