Smoking and stroke: a causative role

Stroke remains the third leading cause of death in most Western countries and is second only to myocardial infarction as a cause of cardiovascular death. Many epidemiological studies have established cigarette smoking as an important risk factor for stroke. Until recently, however, this relation was based on observational studies and the effects of smoking were thought to be synergistic with hypertension, diabetes mellitus, glucose intolerance, age, hypercholesterolaemia, and pre-existing cardiovascular disease.¹ Now we have definite evidence that smoking itself has a direct causal effect on stroke. 

The relation between smoking and atherosclerosis was observed as early as 1908 by Buerger, who noted severe distal ischaemia among young male addicted smokers.² The earliest report associating stroke and extracranial arterial disease is credited to Gowers, who in 1875 showed left carotid artery occlusion in a patient with right hemiplegia and loss of sight in the left eye.³

The Framingham Heart Study was among the first to assess these the relation of smoking to type of stroke, number of cigarettes smoked, and the effect of stopping.⁴ It concluded that smoking made a significant independent contribution to the risk of stroke generally and to brain infarction specifically. The relative risk of stroke in heavy smokers (>40 cigarettes/day) was twice that of light smokers (<10 cigarettes/day), and the risk of stroke increased with the number of cigarettes smoked; cessation lowered the relative risk ratio to that of a non-smoker.⁴ This reduction in risk ratio was significant by two years after stopping and had reached the level of a non-smoker at five years.⁴ In a meta-analysis of 32 separate studies, Shinton and Beevers showed that cigarette smoking independently contributed to the incidence of stroke: the greatest risk was of subarachnoid haemorrhage, followed by cerebral infarction.⁵

Heavy smokers have a relative risk of stroke 2-4 times greater than non-smokers.¹,⁴ The large cohort study of 22 071 US male physicians showed that heavy smokers (>20 cigarettes/day) had a relative risk of total non-fatal stroke of 2.71 and of fatal stroke of 1.46 (P<0.05).⁶ The British Regional Heart Study showed a relative risk of of 3.7 in all current smokers.¹

Howard et al showed increased thickness of the intima-media wall of the carotid artery in smokers compared with non-smokers, particularly among people aged over 60.⁷ Differences in mean maximum intima-media wall thickness in the internal carotid artery between current and non-smokers were greater than the change expected over 10 years for a person who has never smoked.⁸

The association between the number of cigarettes smoked and the increase in the risk of stroke remains inconclusive. Some authors suggest a linear relation, particularly in smokers of more than 20 cigarettes a day and older people.⁴,⁶,^7–9 A dose-response relation between pack years of smoking and carotid artery disease, measured by increased carotid artery intima-media wall thickness, has been shown.⁷,⁸ An association may exist between passive smoking and intimal hyperplasia, potentially increasing the risk of stroke.⁷,¹⁰ Although the dose-response relation is unclear, stopping smoking does reduce the incidence of stroke. Both the Framingham Heart Study⁴ and the Nurses Health Study¹¹ showed a normalised risk ratio five years after cessation. Also after five years there was no further benefit.¹ Tell et al, however, showed that risk reduction was dependent on the quantity of cigarettes smoked before stopping: light smokers (<20 cigarettes/day) reverted to normal values, but heavy smokers retained twice the incidence of stroke as non-smokers.⁸ Secondary pipe or cigar smokers still have an increased risk similar to that of light smokers,¹ so switching to a pipe or cigars confers little benefit.

Former smokers have a decreased prevalence of clinically significant (>50%) internal carotid artery stenosis (7.3%) than current smokers (9.5%),⁸ this difference being greatest in older people.⁷ No significant relation exists between carotid artery wall thickness and years since quitting smoking.

The relative risk of stroke among hypertensive smokers is five times that among normotensive smokers, but 20 times that of normotensive non-smokers. Pharmacological treatment of hypertension in mildly hypertensive smokers is much less effective in reducing the incidence of smoke than in mildly hypertensive non-smokers, supporting smoking as a causal agent.¹² The greatest benefit in stopping smoking is among hypertensive heavy smokers.⁴,⁷

Cigarette smoking is thus a definite independent risk factor for stroke, particularly ischaemic stroke. The mechanisms are poorly understood but may be associated with raised fibrinogen levels, increased packed cell volume, decreased macrophage activity, or changes in lipid biochemistry promoting atherosclerosis. The evidence for a causal association between cigarette smoking and extracranial carotid atherosclerosis is abundant. All smokers who stop smoking will benefit from reducing their risk of ischaemic stroke, irrespective of the degree of previous exposure to smoking.