Online newspaper of professor Yasser Metwally

The author: Professor Yasser Metwally

http://yassermetwally.com

MANAGEMENT OF RISK FACTORS

Stroke is the third leading cause of death in the United States and was the underlying cause of 150,300 deaths in 1988. 3 Stroke was also a major source of disability in the elderly, often leading to institutionalization. The American Heart Association estimates that there were 2,020,000 stroke survivors in the United States in 1986, many of whom required chronic care. 3 Although medical and surgical therapies for impending or recent-onset stroke must be pursued, prevention is likely to be the most effective strategy in reducing the ravages of cerebrovascular disease. The development of atherosclerotic disease and stroke has been related to a number of host and environmental factors delineated in recent years, chiefly through prospective epidemiologic study. The relative impact of each of these risk factors is becoming clearer, and controlled clinical trials have demonstrated the efficacy of risk factor modification in stroke prevention.

Utilizing Framingham Study data, it is possible to examine the determinants, incidence, and manifestations of stroke as they evolved over 36 years of follow-up. Assessment of risk factors that were measured systematically and prospectively many years prior to the appearance of clinical disease provides the least distorted picture of the evolution of these host and environmental factors into clinical disease.

INCIDENCE OF STROKE

The American Heart Association estimates that there were 500,000 new cases of stroke in the United States in 1986. 3 It is further estimated that From the Department of Neurology (PAW), Boston University School of Medicine;Department of Public Health (Epidemiology and Biostatistics) (PAW, RBD), Boston University School of Public Health; Department of Mathematics (AJB, RBD), Statistical Consulting Unit, Boston University; and Section of Preventive Medicine and Epidemiology (PAW), Evans Memorial Department of Clinical Research and Department of Medicine, University Hospital, Boston, Massachusetts approximately 400,000 stroke patients are discharged annually from acute-care hospitals in the United States, three fourths after an initial stroke and the remainder following recurrence. The predisposing factors for cardiovascular disease including stroke have been under study in the general population sample at Framingham, Massachusetts since 1950. These results are based on 36 years of follow-up of 5070 men and women, aged 30 to 62 years, who were free of cardiovascular disease (CVD) at entry into the study and who have undergone examination every 2 years since then.

Neurologic evaluation at the time of the stroke has been made since 1968. During the past 10 years, at least one CT scan of the brain was performed on approximately 85% of cases and stroke type was confirmed by CT scan in 54% of all strokes. Over 36 years, since the study began, the neurologic deficit of the stroke was confirmed by a Framingham Study neurologist in 40% of cases; since 1981, when surveillance was intensified, nearly two thirds of stroke cases were confirmed in this way. The in-hospital assessment by one of the study neurologists has served to document the stroke and determine stroke subtype as well as to differentiate stroke from other neurologic disease. Utilizing these data, it was usually possible to determine if the stroke mechanism was hemorrhage or infarction and to distinguish subarachnoid from intraparenchymatous hemorrhage. Diagnosis of lacunar infarction was based on clinical and CT scan findings, whereas criteria for embolic infarction required a definite cardiac source for embolism. Differentiation of extracranial from intracranial cerebral infarction was made on clinical grounds including noninvasive carotid studies; angiography was performed by the study subjects personal physicians rather infrequently. Follow-up of the population has been satisfactory; less than 10% were completely lost to follow-up for death after 36 years.

During 36 years of follow-up in the Framingham Study, there were 693 cases of stroke and transient ischemic attack (TIA). The relative frequency of stroke by type shows no substantial difference in stroke manifestation by sex . Atherothrombotic brain infarction (ABI), including infarction secondary to large-vessel atherothrombosis as well as lacunar infarction, occurred most frequently, comprising 44% of strokes. TIA only accounted for approximately 20% of all stroke events and was more frequent in men, whereas cerebral embolism was marginally more frequent in women (23%) than men (18%). Intracranial hemorrhage accounted for 11% of all stroke events; subarachnoid hemorrhage was slightly more frequent than intraparenchymatous bleeding.

The average annual incidence of all strokes combined increased with age and doubled in each successive decade ; this pattern also held true for brain infarction alone. Incidence of all strokes generally and ABI specifically was approximately 30% higher in men than in women. It should be noted that these rates were derived using as denominator a population at risk which was free all cardiovascular disease at entry and is different from that used in some prior publications.

RISK FACTORS FOR STROKE

• Atherogenic Host Factors

Identification of a risk factor and awareness of its relative importance and its interaction with other precursors may yield important clues concerning pathogenesis of cerebrovascular disease and thereby lead to stroke prevention. Assessment of the importance of each of the major atherogenic risk factors was made specifically in that stroke type resulting most directly from atherothrombosis, ABI. The relative impact of each of these risk factors can be assessed in each sex by comparing the size of the multivariate regression coefficients . These coefficients were standardized for the varied units and scales and thereby take into account the contribution made by other risk factors. The larger the standardized multivariate regression coefficient, the greater the impact on ABI incidence. Statistical significance indicates that the risk factor made a significant independent contribution to risk in that age category even after the other cardiovascular risk factors were taken into account. These risk factors included definite hypertension, systolic blood pressure, diastolic blood pressure, left ventricular hypertrophy by ECG, cigarette smoking, total serum cholesterol, glucose intolerance, Metropolitan relative weight, and hematocrit. In addition, comparing the size of the coefficients in men and women and in younger and older age groups permits a clearer picture of the relative impact of each risk factor on brain infarction incidence.

• Hypertension.

Hypertension is the paramount risk factor for stroke. The standardized multivariate regression coefficient for definite hypertension, as a categorical variable for levels of blood pressure 160/95 mm Hg or greater, is large in both men and women, denoting the powerful and independent contribution of definite hypertension to incidence of brain infarction. It is noteworthy that the coefficient for hypertension is smaller in the older age group , corresponding to a reduced but still highly statistically significant effect.

• • Systolic Versus Diastolic Blood Pressure Level.

Both systolic and diastolic blood pressure levels are strongly and independently related to stroke incidence. Both components of blood pressure have the largest coefficients even in the oldest age group when compared with the other risk factors. By comparing standardized coefficients , however, it is apparent the systolic blood pressure coefficients are only moderately reduced in the oldest age group, in which they are definitely larger than the diastolic in both men and women. These data contradict the notion that increased blood pressure is not an important risk factor for stroke in the elderly. Further, with respect to stroke, there is no evidence that women tolerate hypertension better than men or that diastolic elevations are more important than systolic pressures.

• • Isolated Systolic Hypertension.

With advancing age, there is a disproportionate rise in systolic blood pressure while the diastolic pressure levels off and then begins to decline. In the elderly, this isolated elevation of systolic pressure becomes highly prevalent. Isolated systolic hypertension-systolic pressure greater than 160 mm Hg and diastolic pressure less than 90 or 95 mm Hg-is present in approximately 20% of men and 30% of women above age 75. 15 This isolated elevation of the systolic pressure results from decreased arterial wall elasticity, which is a consequence of arteriosclerotic changes and was considered to be a consequence rather than a precursor of cardiovascular disease. However, it has been demonstrated in the Framingham Study and other epidemiologic studies that stroke and cardiovascular disease incidence are significantly increased in persons with isolated systolic hypertension. 5, 19 Risk is proportionately related to the level of systolic pressure even after diastolic pressure, age, and digital pulse-wave configuration (an index of arterial rigidity) are taken into account. It is apparent from these 36-year follow-up data from Framingham that incidence of stroke generally and, to a lesser extent, brain infarction is increased in the elderly with isolated systolic hypertension. The recent report of a striking beneficial effect of blood pressure reduction in isolated systolic hypertensives gives these data added significances 38 and is emphasized below.

• Heart Disease and Impaired Cardiac Function.

Cardiac disease and impaired cardiac function are not risk factors but rather represent a group of disorders that predispose to stroke. Although hypertension is the pre-eminent risk factor for stroke of all types, at any blood pressure level persons with impaired cardiac function, occult or overt, have a significantly increased stroke risk. 49 Pertinent common cardiac abnormalities include prior coronary heart disease (CHD), cardiac failure, ECG abnormalities (particularly left ventricular hypertrophy), and atrial fibrillation. Cardiac disease is an important precursor of stroke and is dealt with in detail in several other articles. In Framingham, CHD was ascertained prospectively on biennial examination as well as by monitoring hospitalizations; all manifestations, including uncomplicated angina pectoris and clinically silent as well as overt myocardial infarctions, were significantly related to increased stroke risk. 20

• • Atrial Fibrillation.

In association with rheumatic heart disease and mitral stenosis, atrial fibrillation is acknowledged to predispose to stroke. In recent years, chronic atrial fibrillation without valvular heart disease, previously considered to be innocuous, has been associated with a more than fivefold increased incidence of stroke. It is also the most prevalent cardiac arrhythmia in the elderly. In the Framingham Study atrial fibrillation incidence more than doubled in successive decades and rose from 0.2 per 1000 for ages 30 to 39 to 39.0 per 1000 for ages 80 to 89 years. Atrial fibrillation was particularly important in the elderly because the proportion of strokes associated with this arrhythmia increased steadily with age, rising from 6.7% for ages 50 to 59 years to 36.2% for ages 80 to 89 years. 48 Atrial fibrillation was present in one fifth of all initial stroke events in the eighth decade of life and approximately one third of those in the ninth decade of life. The stroke was not a consequence of the often associated CHD or cardiac failure because relative risk of stroke rose with age for atrial fibrillation, whereas relative risk of stroke attributable to cardiac failure, CHD, and hypertension declined with age. 49

Although most persons with nonrheumatic atrial fibrillation do not sustain a stroke and stroke prevention with chronic warfarin anticoagulation is not innocuous, recent clinical trial findings strongly support low-intensity warfarin therapy with greater than 85% effectiveness in stroke prevention when prothrombin time prolongation was in the therapeutic range.

• • Left Ventricular Hypertrophy by ECG.

Left ventricular hypertrophy by ECG, probably a result of prolonged and severe hypertension, increases in prevalence with age and blood pressure. Risk of ABI increased more than fourfold in persons with this abnormal ECG pattern. This excess risk persisted even after the influence of other atherogenic precursors, including blood pressure and age, were taken into account .

• Blood Lipids.

Total serum cholesterol is significantly and independently related to the development of coronary heart disease in men and women, but the impact seems to diminish beyond age 55, particularly in men. However, when the components of cholesterol-high, low, and very low density lipoprotein cholesterol are related to incidence of CHD, a relationship re-emerges even in persons over age 55. High density lipoprotein (HDL) cholesterol has an inverse association and low density lipoprotein (LDL) cholesterol a direct relationship to incidence of CHD. The relation of total serum cholesterol and lipoprotein cholesterol fractions to the occurrence of stroke or ABI is far less clear or consistent (.

Serum lipid levels have been related to carotid artery atherosclerosis in a number of studies using the ultrasonographic evidence of extracranial carotid artery atherosclerosis or internal carotid artery wall thickness as an indicator of atherosclerosis. 30, 37 It seems clear that atherosclerosis of the carotid artery, and of the circle of Willis in autopsy studies, are related to levels of blood lipids. On the other hand, the relationship to stroke generally may be obscured by the differing influence of lipids on the varying vascular pathologies underlying stroke. Levels of total serum cholesterol below 180 mg/dL, and particularly below 160 mg/dL, seem to promote intraparenchymatous and perhaps subarachnoid hemorrhage, whereas elevated levels foster large vessel atherothrombosis. There is no apparent influence on lacunar infarcts or on strokes secondary to cerebral embolism.

The high rates of hemorrhage and stroke generally and low rates of CHD in Japanese in Japan and Hawaii have occurred in persons with low total serum cholesterol levels, by western standards, and in the presence of a high prevalence of hypertension. 39 This tendency for low levels of total cholesterol to promote intracerebral hemorrhage was well documented in Asians but not in Caucasians until recently. A relationship between high levels of total serum cholesterol and ischemic stroke and low levels of serum cholesterol and intracerebral hemorrhage was found after 6 years of follow-up of predominantly white males from the United States. 13 These data from the Multiple Risk Factor Intervention Trial (MRFIT) related the single cholesterol level of 350,977 men screened for the trial to a death certificate diagnosis of fatal stroke. Despite the limitations of death certificate diagnoses, the findings are quite persuasive as a result of the large sample size and the prospective nature of the study. These data have led some to question the safety and prudence of the recent US cholesterol education program. 13

• Diabetes.

Diabetics are known to have an increased susceptibility to coronary, femoral, and cerebral artery atherosclerosis. Surveys of stroke patients and prospective study have confirmed the increased risk of stroke in diabetics. In the United States, in the period 1976 to 1980, a medical history of stroke was 2.5 to 4 times more common in diabetics than in persons with normal glucose tolerance. 23 In Framingham, peripheral arterial disease with intermittent claudication occurs more than four times as often in diabetics. The coronary and cerebral arteries are also affected but to a lesser extent. 17 For atherothrombotic brain infarction, the impact of glucose intolerance, i.e. physician-diagnosed diabetes, glycosuria, or a blood sugar greater than 150 mg/100 mL, is greater in women than in men and is significant as an independent contributor to incidence only in older women .

• Obesity.

Obese persons have higher blood pressures and higher blood glucose levels, as well as higher levels, of atherogenic serum lipids. On that account alone, the increased prevalence of hypertension and impaired glucose tolerance associated with obesity could be expected to increase stroke incidence. Obesity, expressed as a Metropolitan relative weight that is more than 30% above average, is a significant independent contributor to ABI incidence in men aged 35 to 64 and women 65 to 94 years . Even in the other two age-sex groups, obesity exerts an adverse influence on health status which is probably mediated through elevated blood pressure, impaired glucose tolerance, and other mechanisms. Recent studies have suggested that the pattern of obesity is also important, with central obesity and abdominal deposition of fat more strongly associated with atherosclerotic disease. 9

• Hematocrit.

Some studies, including the Framingham Study, demonstrated a relationship between high normal hematocrit level and increased incidence of cerebral infarction. 16 Confirmation of this relationship has come from an autopsy study of Japanese stroke patients and from several clinical and radiologic studies of patients with stroke. 43, 44 In these 36-year follow-up data, elevated blood hematocrit within the normal range, and generally not pathologically elevated red cell mass, is significantly and independently associated with ABI in men aged 35 to 64, but not in the other three age-sex categories. This significant relationship to ABI persisted even after the confounding effects of cigarette smoking and hypertension were taken into account . The reason for the different impact in the two sexes is unclear. Increased concentration of red cells in combination with high blood fibrinogen levels raises blood Viscosity. This interplay may reach pathologic significance in narrowed small penetrating arteries and in high-grade stenosis of a major cerebral artery. Reduction of high-normal hematocrit by venesection has decreased blood viscosity and correspondingly increased cerebral blood flow.

• Fibrinogen.

Serum fibrinogen has been implicated in atherogenesis and in arterial thrombus formation. A number of epidemiologic studies have shown a substantial and significant independent impact of fibrinogen on cardiovascular disease incidence, including stroke. 46 In this prospective study, fibrinogen in combination with elevated systolic blood pressure, was found to be a potent risk factor for stroke in a sample of 54-year-old Swedish men followed for 13 years. 45 Level of fibrinogen, measured on the tenth biennial examination in Framingham, was significantly related to incidence of cardiovascular disease, including stroke. 18 However, fibrinogen was also positively associated with most of the major risk factors for stroke, including age, hypertensive status, hematocrit level, obesity, and diabetes. There is considerable optimism that further study of fibrinogen and other clotting factors in atherosclerotic cardiovascular disease will yield important insights into mechanisms of disease and potential preventive measures.

• Race.

Japanese have been known to have low rates of CHD and a high prevalence of stroke. A survey of six mainland Chinese cities disclosed a high incidence and prevalence of stroke in mainland Chinese, comparable to those of native Japanese in Japan. 26 The disparity between the stroke and CHD death rates and presumably a similar disparity in incidence rates are usually attributed to the high prevalence of hypertension and the low levels of blood lipids in these racial groups. 35 Cerebrovascular disease was the most frequently certified cause of death in Japan during the decades following World War II, and the mechanism of stroke was most frequently thought to be intraparenchymatous hemorrhage. By 1980, as stroke death rates fell, cancer became the leading cause of death. In Japanese men in Hawaii and San Francisco, deaths attributed to stroke also fell relative to CHD (and cancer) death rates. 35 In 1985, heart disease became the second leading cause of death in Japan (114.8 per 100,000) following cancer (156.4 per 100,000), and stroke fell to the third position (110.6 per 100,000). 14

It is now well established that infarction, not hemorrhage, is the most frequent stroke mechanism in the Japanese and accounts for two thirds of stroke events in this population in both Japan and Hawaii. However, intracerebral hemorrhage does occur several-fold more frequently in Japanese than in US whites or blacks. There is also a difference in the site of the arterial pathology, with a predominance of intracranial disease in native Japanese in contrast to the pattern in white Americans, in whom the extracranial arteries are the focus of most of the atherosclerotic occlusive disease. In Japan, including rural Japan, substantial changes have occurred in the diet since World War II. These include an increase in animal fat and animal protein and a reduction in the amount of sodium chloride in the diet. 39

• Family History of Stroke.

Although family history of stroke is perceived to be an important marker of increased stroke risk, confirmation by epidemiologic study has been lacking. Recently, in a cohort of Swedish men born in 1913, maternal history of death from stroke was significantly related to stroke incidence. 45 Other significant risk factors included hypertension, abdominal pattern of obesity, and fibrinogen level; however, maternal history of fatal stroke was independently related to stroke even after these variables were taken into account.

• Environmental Factors

  • Cigarette smoking

Cigarette smoking,a powerful risk factor for myocardial infarction and sudden death, has been clearly linked to brain infarction as well as to intracerebral and subarachnoid hemorrhage. 6, 51 A similar relationship between cigarette smoking and stroke has been seen in Hawaiian Japanese men after 10 years of follow-up in the Honolulu Heart Study, in which cigarette smoking made a significant independent contribution to cerebral infarction and intracranial hemorrhage risk. 1 In a meta-analysis of 32 separate studies, including those cited above, cigarette smoking was a significant independent contributor to stroke incidence in both sexes and at all ages and was associated with approximately a 50% increased risk overall when compared with nonsmokers. 40

• Oral Contraceptives.

An increased risk of stroke was reported in users of oral contraceptives (OCs), particularly older women (above age 35) and predominantly in those with other cardiovascular risk factors, particularly hyper- tension and cigarette smoking. 41 The relative risk of stroke was estimated to be increased fivefold in OC users and former users, with risk concentrated in cigarette smokers above age 35. The pathogenesis of stroke in OC users is unclear. Arterial occlusion is likely to be due to thrombotic occlusion and not atherosclerosis, and it is known that clotting is enhanced by the OC-induced increased platelet aggregability and by its alteration of clotting factors to favor thrombogenesis. In young women with unexplained ischemic stroke, use of OCs is presumed to be the cause of the infarct; OC use was the cause in 10% of a series of carefully studied patients. In the older case-control studies of OC users (such reports are far less evident in recent years), “thromboembolism” was the type of stroke said to be most frequent in OC users. However, these series often comprised women with transient episodes of neurologic dysfunction (TIAS) and examples of stroke were ascribed to “ill-defined and uncertain” cause. Risk of stroke was highest in women who took OCs containing higher levels of estrogen; the lower levels of estrogen in the newer OC formulations seem to have substantially reduced the risk. 42

Of particular interest is the interaction between OCs, cigarette smoking, and subarachnoid hemorrhage. 27 Prospective observation of over 40,000 women, half of whom were taking OCs, showed an increased risk of fatal subarachnoid hemorrhage (not cerebral infarction) in women taking OCS. 36 Risk was increased fourfold in cigarette smokers above age 35, with most cases confined to this group.

• Alcohol Consumption.

As in myocardial infarction, impact of alcohol consumption on stroke risk is related to the amount of alcohol consumed. 21 Heavy alcohol use, either habitual daily heavy alcohol consumption or binge drinking, seems to be related to an excess of stroke and stroke deaths. Light or moderate alcohol consumption, on the other hand, is convincingly associated with a reduced incidence of CHD. Light and moderate alcohol use tends to raise the HDL cholesterol and may be associated with a reduction in CHD incidence, whereas high levels of alcohol intake are linked to hypertension and hypertriglyceridemia and may, in this way, be associated with an increased rate of CHD.

The relationship of alcohol consumption to stroke occurrence is less clearly elucidated. Available evidence rather uniformly demonstrated an adverse effect of heavy alcohol consumption on stroke occurrence. Stroke incidence has been positively associated with alcohol consumption in prospective studies in Yugoslavia and in Japanese men in Hawaii. 8, 22 There was a powerful close- response relationship between alcohol consumption in the Honolulu Heart Study of men of Japanese ancestory, 8 and incidence of intracerebral and subarachnoid hemorrhage, even after taking other pertinent risk factors, particularly blood pressure, into account. There was no significant relationship to thromboembolic stroke, however.

Data from the Framingham Study also suggest an increased incidence of brain infarction and stroke with increased levels of alcohol use, but only in men. There are a number of mechanisms by which heavy alcohol consumption may predispose to stroke. 11 Cigarette smoking is more frequent in heavy drinkers, and there is attendant hemoconcentration. Alcohol and cigarette smoking have been shown to increase blood hematocrit and viscosity, and rebound thrombocytosis during abstinence has been observed. Cardiac rhythm disturbances occur with alcohol intoxication, and acute inebriety has been suggested to be a precipitating factor in stroke in young people, in thrombotic stroke, and in subarachnoid hemorrhage.

• Physical Activity.

Leisure-time and work-associated vigorous physical activity has been linked to lower CHD incidence. More physically active longshoremen had lower rates of myocardial infarction but had no reduction in stroke incidence. 32 In a study of 17,000 former Harvard students, those who were more physically active had about half the risk of fatal CHD and one third the mortality rate of their least active fellow alumni. 33 No such data exist demonstrating a reduction in stroke occurrence from vigorous habitual leisure- time physical activity.

Vigorous exercise may exert a beneficial influence on risk factors for atherosclerotic disease by reducing elevated blood pressure as a result of weight loss and by reducing the pulse rate, raising the HDL cholesterol and lowering the LDL cholesterol, improving glucose tolerance, and promoting a life style conducive to favorably changing detrimental health habits such as cigarette smoking. 24,25 These salutary effects have not been uniformly demonstrated to result in reduced stroke incidence. Furthermore, there are substantial limitations to studies of physical activity and disease, the principal potential pitfall being the self-selection of different levels of activity at work or leisure.

STROKE PREVENTION THROUGH RISK FACTOR MANAGEMENT

There has been rapid decline in death rates from stroke in the United States and in most other industrialized nations since 1968. 21 In the United States this decline of more than 50% in mortality rates in a 20-year span supports the notion that modifiable environmental influences are operating in stroke and cardiovascular disease occurrence. At least part of the decline results from a reduction in the incidence and severity of stroke which most attribute to improved detection and treatment of hypertension. 10 Prevention of stroke, and perhaps stroke recurrence, may be accomplished by reduction of elevated blood pressure, prevention and treatment of predisposing cardiac diseases, and probably by cessation of cigarette smoking. Although CHD incidence and recurrence can be reduced by cholesterol lowering, there is little evidence that such an effort would reduce stroke directly. Of course, CHD is a major precursor of stroke and is the principal cause of death of stroke and TIA survivors, and on that account CHD prevention is certainly worthwhile.

• Control of Hypertension and Stroke Prevention

Following clinical observation of patients with treated and untreated hypertension with prospective epidemiologic study, it is apparent that level of blood pressure is related to incidence of stroke for severe, moderate, and even mild hypertensives. 29 A combined analysis of nine major prospective studies, including 420,000 individuals with a mean 10-year follow-up, shows clear evidence of a graded relationship between diastolic pressure and stroke and CHD incidence. There is no threshold level below which risk gradients are flat. For every 7.5 mm Hg diastolic pressure increase, there is a 46% increase in stroke incidence and a 29% increase in CHD. Relating these findings from prospective observational study to randomized trials of blood pressure reduction demonstrates that treatment prevented stroke. The findings should put to rest the concern that reduction of blood pressure in hypertensives serves to precipitate stroke. From a statistical analysis of 14 treatment trials with a total of 37,000 hypertensive subjects, it is clear that reduction of blood pressure in hypertensives reduced stroke incidence. 7 There was an average blood pressure reduction of 5.8 mm Hg and a corresponding reduction in stroke incidence of 42%. This observed reduction in stroke closely approximated that expected on the basis of prospective observational studies.

Unfortunately, the 14% decrease in CHD rates was considerably lower than expected. The explanation for the lesser effect on CHD incidence is being sought. It may relate to the adverse influence on electrolyte, glucose, and lipid metabolism of the thiazide diuretics that were the mainstay of treatment in most of the trials. 28 It is important to note that these substantial and significant reductions in stroke incidence occurred during the course of the trials. In these studies, the duration of blood pressure reduction was brief, from 2 to 5 years, suggesting interruption of a precipitating factor rather than interference with atherogenesis. Presumably, more prolonged blood pressure control would have both effects.

Emphasis has been placed on the diastolic component in virtually all treatment trials, although stroke risk is clearly no less directly related to systolic pressure levels . In the elderly, in whom isolated elevation of the systolic pressure is common, treatment was thought to be ineffective in reducing pressure, hazardous in terms of side effects, and unwarranted on the basis of available epidemiologic data. In 4736 persons above age 60 with systolic blood pressure levels above 160 mm Hg and diastolic pressures below 90 mm Hg, blood pressure reduction was associated with a 36% reduction in stroke and a 27% reduction in myocardial infarction and coronary death after 4.5 years of follow-up. 28 These findings have enormous importance because two thirds of all individuals with hypertension between the ages of 65 and 89 years have isolated systolic hypertension. Most strokes occur in this age group. 47 It is clear from the SHEP Trial and from the European Working Party on Hypertension in the Elderly (EWPHE) study, that antihypertensive medication is well tolerated by the elderly. 4 SHEP demonstrated that reduction of pressure was accomplished with relative ease (approximately half were controlled with chlorthalidone alone) and was well tolerated, as evidenced by a 90% compliance rate in the active treatment group at 5 years. Because increased blood pressure is the most powerful risk factor for stroke and because the benefits of treatment occur so promptly, control of increased blood pressure-systolic as well as diastolic levels-is the cornerstone of stroke prevention.

• Cessation of Cigarette Smoking

Based on data from the Nurses Health Study and from the Framingham Study, it seems clear that stopping smoking is followed by a reduction in stroke risk in a remarkably short time. 6, 51 Risk of CHD decreases by approximately 50% within 1 year of smoking cessation and reaches the level of those who never smoked within 5 years. In Framingham, in both men and women, risk of stroke in former cigarette smokers did not differ from that of persons who never smoked by the end of 5 years. There was no age effect, suggesting that cigarette smoking exerted a precipitating effect on stroke, myocardial infarction, and sudden death which was reversed with smoking cessation. Since smoking confers an increase in stroke risk of 40% in men and 60% in women, after all other pertinent risk factors have been taken into account, cessation may be expected to significantly reduce risk of stroke.

• Prevention and Treatment of Heart Disease Including Atrial Fibrillation

Since CHD, cardiac failure, and atrial fibrillation predispose to stroke, prevention of these cardiovascular contributors can be anticipated to reduce incidence of stroke. 49 On the basis of current knowledge of the epidemiology of cardiac failure, prevention of obesity and treatment of hypertension may be beneficial. Reduction of CHD risk requires, in addition to hypertension control and smoking cessation, dietary or pharmacologic treatment to reduce elevated total and LDL cholesterol and increase the HDL cholesterol fraction. Prevention of atrial fibrillation might best be accomplished by preventing the appearance of its major precursor, which is heart disease.

• Identification of High-risk Candidates for Stroke Prevention

Each physician can identify “prime” candidates for stroke among his asymptomatic patients. Control of severe and moderately severe hypertension definitely prevents stroke. Patients with these levels of blood pressure require vigorous and sustained therapy to maintain normotension. However, 70% of hypertensives have borderline elevations of blood pressure. Treatment of this large population with mild and borderline blood pressure elevation with drugs would be an extraordinary therapeutic and financial effort. Although the rewards of such an effort may be correspondingly great, it seems more likely that many persons whose risk of developing atherosclerotic disease is low would be exposed to costly drugs with dangerous or unpleasant side effects. What is needed is a way of selecting persons at substantially increased risk of stroke and other cardiovascular disease for intensive preventive intervention, including pharmacologic treatment of hypertension. For the remainder, whose risk of stroke is average or only mildly increased, hygienic measures should be prescribed. These include weight loss, reduction of salt (as well as calories and fat) in the diet, cigarette smoking cessation, and moderate physical activity, all measures that can be advocated for most people.

A stroke risk profile has been developed, utilizing data from 36 years of follow-up from Framingham, which allows the physician to determine a patient’s probability of stroke. 50 This can be accomplished based on information collected during the course of taking a comprehensive medical history and conducting a physical examination, plus obtaining an ECG. Using a table that is sex-specific, probability is determined by a point system depending on age, systolic blood pressure, antihypertensive therapy use, presence of diabetes, cigarette smoking, history of cardiovascular disease (CHD or cardiac failure), and ECG abnormalities (left ventricular hypertrophy or atrial fibrillation). Risk is distributed over a wide range and permits the physician to rapidly relate a particular patient’s probability of stroke to that of an average person of the same age and sex.

The stroke risk profile helps the physician to identify which borderline hypertensives warrant pharmacologic treatment by virtue of an increased probability of stroke, usually attributable to the presence of several other risk factor abnormalities. By restricting drug treatment to persons with a systolic blood pressure level who have two or more other risk factor abnormalities (not including age and male sex), it is possible to identify a group consisting of 22% of men and 14% of women at high risk. In this group approximately 40% of strokes occur within the subsequent 10 years. 50

Clearly there are other situations not considered here when a patient can be identified to be at substantially increased risk of stroke: recent TIA, recent onset atrial fibrillation, recent myocardial infarction, cardiac surgery, and others that are dealt with elsewhere.

---

References

1. Abbott RD, Yin Yin MA, Reed DM, Yano K: Risk of stroke in male cigarette smokers. N Engl J Med 315:717-720, 1986

2. Adams HP, Butler Mj, Biller J, Toffol Gj: Nonhemorrhagic cerebral infarction in young adults. Arch Neurol 43:793-796, 1986

3. American Heart Association: 1991 Heart and Stroke Facts. Dallas, American Heart Association, 1991

4. Amery A, Birkenhager W, Brixko P, et al: Mortality and morbidity results from the European Working Party on High Blood Pressure in the Elderly Trial. Lancet 1:1349- 1354, 1985

5. Colandrea MA, Friedman GD, Nichaman MZ, et al: Systolic hypertension in the elderly: An epidemiologic assessment. Circulation 41:239-245, 1970

6. Colditz GA, Bonita R, Stampfer Mj, et al: Cigarette smoking and risk of stroke in middle-aged women. N Engl J Med 318:937-941, 1988

7. Collins R, Peto R, MacMahon S, et al: Blood pressure, stroke and coronary heart disease: Part 11. Effects of short-term reductions in blood pressure and overview of the unconfounded randomized drug trials in an epidemiological context. Lancet 1:827-838, 1990

8. Donahue RP, Abbott RD, Reed DM, Yano K: Alcohol and hemorrhagic stroke: The Honolulu Heart Program. JAMA 255:2311-2314, 1986

9. Folsom AR, Prineas Rj, Kaye SA, Munger RG: Incidence of hypertension and stroke in relation to body fat distribution and other risk factors in older women. Stroke 21:701-706, 1990

10. Garraway WM, Whisnant JP: The changing pattern of hypertension and the declining incidence of stroke. JAMA 258:214-217, 1987

11. Gorelick PB: The status of alcohol as a risk factor for stroke. Stroke 12:1607-1610, 1989

12. Hennekens CH: Alcohol. In Kaplan NM, Stamler J (eds): Prevention of Coronary Heart Disease, Practical Management of the Risk Factors. Philadelphia, WB Saunders, 1983, pp 130-138

13. Iso H, Jacobs DR, Wentworth D, et al: Serum cholesterol levels and six-year mortality from stroke in 350,977 men screened for the multiple risk factor intervention trial. N Engl J Med 320:904-910, 1989

14. Japanese Health and Welfare Ministry. AMA Medical News 17:12, 1986

15. Kannel WB, Gordon T: Evaluation of cardiovascular risk in the elderly: The Framingham Study. Bull NY Acad Med 54:573-591, 1978

16. Kannel WB, Gordon T, Wolf PA, et al: Hemoglobin and the risk of cerebral infarction: The Framingham Study. Stroke 3:409-419, 1972

17. Kannel WB, McGe DL: Diabetes and cardiovascular disease: The Framingham Study. JAMA 241:2035-2038, 1979

18. Kannel WB, Wolf PA, Castelli WP, et al: Fibrinogen and risk of cardiovascular disease: The Framingham Study. JAMA 258:1183-1186, 1987

19. Kannel WB, Wolf PA, McGee DL, et al: Systolic blood pressure, arterial rigidity, and risk of stroke: The Framingham Study. JAMA 245:1225-1229, 1981

20. Kannel WB, Wolf PA, Verter J: Manifestations of coronary disease predisposing to stroke: The Framingham Study. JAMA 250:2942-2946, 1983

21. Kiag Mj, Whelton PK, Seidler AJ: Decline in US stroke mortality: Demographic trends and antihypertensive treatment. Stroke 20:14-21, 1989

22. Kozarevic DJ, McGee D, Vojvodic N, et al: Frequency of alcohol consumption and morbidity and mortality: The Yugoslavia Cardiovascular Disease Study. Lancet 1:613- 616, 1980

23. Kuller LH, Dorman JS, Wolf PA- Cerebrovascular diseases and diabetes. In Diabetes in America, Diabetes Data Compiled for 1984. National Diabetes Data Group, Department of Health and Human Services. NIH Publication No. 85-1468, August, XVIII:1-18, 1985

24. Leon AS, Blackburn H: Physical Inactivity. In Kaplan NM, Stamler J (eds): Prevention of Coronary Heart Disease, Practical Management of the Risk Factors. Philadelphia, WB Saunders, 1983, pp 86-97

25. Leon AS, Connett J, Jacobs DR, Rauramaa R: Leisure-time physical activity levels and risk of coronary heart disease and death: The Multiple Risk Factor Intervention Trial. JAMA 258:2388-2395, 1987

26. Li S, Schoenberg BS, Wang C, et al: Cerebrovascular disease in the People’s Republic of China: Epidemiologic and clinical features. Neurology (Cleveland) 35:1708-1713, 1985

27. Longstreth WT, Koepsell TD, Yerby MS, van Belle G: Risk factors for subarachnoid hemorrhage. Stroke 16:377-385, 1985

28. MacMahon S, Cutler JA, Stamler J: Antihypertensive drug treatment: Potential, expected and observed effects on stroke and on coronary heart disease. Hypertension 13(Suppl 1):I45-150, 1989

29. MacMahon S, Peto R, Cutler J, et al: Blood pressure, stroke and coronary heart disease: Part 1. Effects of prolonged differences in blood pressure. Evidence from nine prospective observational studies corrected for regression dilution bias. Lancet 335:765-774, 1990

30. O’Leary DH, Anderson KM, Wolf PA, et al: Cholesterol and carotid atherosclerosis in older persons: The Framingham Study. Ann Epidemiol, in press

31. Oliver MF: Might treatment of hypercholesterolaemia increase non-cardiac mortality? Lancet 1:1529-1531, 1991

32. Paffenbarger RS, Laughlin ME, Gina AS, et al: Work activity of longshoremen as related to death from coronary heart disease and stroke. N Engl J Med 282:1109- 1115, 1970

33. Paffenbarger RS, Wing AL, Hyde RT: Physical activity as an index of heart attack risk in college alumni. Am J Epidemiol 108:161-175, 1978

34. Pearson TC, Thomas DJ: Physiological and pharmacological factors influencing blood viscosity and cerebral blood flow. In Tognomi G, Gerattinin S (eds): Drug Treatment and Prevention in Cerebrovascular Disorders. Amsterdam, Elsevier-North Holland, 1979, p 33

35. Reed DM: The paradox of high risk of stroke in populations with low risk of coronary heart disease. Am J Epidemiol 131:579-588, 1990

36. Royal College of General Practitioners’ Oral Contraception Study: Further analyses of mortality in oral contraceptive users. Lancet 1:541-546, 1981

37. Salonen R, Seppanen K, Rauramaa R, Salonen JT: Prevalence of carotid atherosclerosis and serum cholesterol levels in eastern Finland. Atherosclerosis 8:788-792, 1988

38. SHEP Cooperative Research Group: Prevention of stroke by antihypertensive drug, treatment in older persons with isolated systolic hypertension: Final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA 265:3255-3264, 1991

39. Shimamoto T, Komachi Y, Inada H, et al: Trends for coronary heart disease and stroke and their risk factors in Japan. Circulation 79:503-515, 1989

40. Shinton R, Beevers G: Meta-analysis of relation between cigarette smoking and stroke. Br Med J 298:789-794, 1989

41. Stadel BV: Oral contraceptives and cardiovascular disease. N Engl J Med 288:672- 677, 1981

42. Stampfer Mj, Willett WC, Colditz GA, et al: A prospective study of past use of oral contraceptive agents and risk of cardiovascular diseases. N Engl J Med 319:1313- 1317, 1988

43. Thomas DJ, Marshal J, Ross Russell RW, et al: Effect of haematocrit on cerebral blood flow in man. Lancet 2:1163-1164, 1966

44. Tohgi H, Yamanouchi H, Murakami M, et al: Importance of the hematocrit as a risk factor in cerebral infarction. Stroke 9:369-374, 1978

45. Welin 1, Svardsudd K, Wilhelmsen L, et al: Analysis of risk factors for stroke in a cohort of men born in 1913. N Engl J Med 317:521-526, 1987

46. Wilheimsen L, Svardsudd K, Korban-Bengsten K, et al: Fibrinogen as a risk factor for stroke and myocardial infarction. N Engl J Med 311:501-505, 1984

47. Wilking SVB, Belanger A, Kannel WB, et al: Determinants of isolated systolic hypertension. JAMA 260:3451-3455, 1988

48. Wolf PA, Abbott RD, Kannel WB: Atrial fibrillation: A major contributor to stroke in the elderly: The Framingham Study. Arch Intern Med 147:1561-1564, 1987

49. Wolf PA, Abbott RD, Kannel WB: Atrial fibrillation, an independent risk factor for stroke: The Framingham Study. Stroke 22:983-988, 1991

50. Wolf PA, D’Agostino RB, Belanger Aj, Kannel WB: Probability of stroke: A risk profile from the Framingham Study. Stroke 22:312-318, 1991

51. Wolf PA, D’Agostino RB, Kannel WB, et al: Cigarette smoking as a risk factor for stroke: The Framingham Study. JAMA 259:1025-1029, 1988