Stroke Prevention
Show Me The Evidence

The prevention of stroke, with its attendant costs, both financial and personal, is the goal of most family physicians, but is there good evidence that modification of risk factors will reduce stroke risk? In this article we'll look at the evidence for the following interventions: treatment of hyperlipidemia, smoking cessation, antiplatelet therapy, treatment of hypertension and anticoagulation for atrial fibrillation.

Does aggressive treatment of dyslipidemia decrease the risk of stroke?
This issue has been looked at indirectly in multiple placebo-controlled trials, as a pre-specified secondary end point. Perhaps the best place to demonstrate this is to look at the MRC/BHF study, commonly known as the Heart Protection Study.¹ This is the largest trial ever done on lipid lowering. It included 20,536 people aged 40 to 80 with coronary disease, other occlusive arterial disease or diabetes with a non fasting cholesterol >3.5 mm/l. There were 6,793 subjects with a starting LDL < 3mm/l. They were randomized to 40 mg of simvastatin or a matching placebo. Follow up was for an average of 5 years. 4.3% of the simvastatin group vs. 5.7% of the placebo group suffered a stroke. (RR 75%, RRR 25%, ARR 1.4% NNT 71). This small but consistent effect is seen in multiple trials. There were no excess hemorrhagic strokes in the treatment arm.

Does smoking increase the risk of stroke? What evidence do we have?
Clearly we do not have the best Level 1 evidence here as no randomized controlled trials have been done on this subject. In issues of harm, RCTs are impossible to do, as it is ethically impossible to randomize people to a smoking arm! So we have to look at the next best evidence, a prospective cohort trial, which systematically follows a large group of healthy people until they get ill with the disease of interest, and then tries to elucidate the risk factors for that outcome.

The Framingham Study began soon after WWII, is still ongoing, and is the longest running prospective study of this type. In 1988,² it reported on cigarette smoking as a risk factor for stroke. It looked at 4,255 men and women, 36 to 68 years old and free of cerebrovascular disease. During a 26-year follow up 459 strokes occurred. Multivariate analysis using Cox proportional hazard modeling, demonstrated cigarette smoking as an independent risk factor for all strokes in general, and thrombotic strokes in particular. The risk of stroke increased as the number of cigarettes smoked increased. The RR was double in heavy smokers (> 40 cigarettes/ day) compared to light smokers (< 10 cigarettes/day). Lapsed smokers retained their increased risk for the first 2 years, which then gradually decreased over the next 3 years. By 5 years the risk dropped to the level of non-smokers. Regardless of smoking status or sex, hypertension doubled the risk of stroke.

In 1989 a case controlled study³ (a less robust study design due to its inherent bias) looked at 621 patients with stroke and 573 controls without stroke. The authors estimated an increased relative risk of 1.5 for every 10 cigarettes consumed daily, both in men and women.

Antiplatelet Medications

The use of antiplatelet medications in the secondary prevention of stroke has long been established. Various drugs and combinations of drugs have been studied over the last 20 years. We will look at the evidence for ASA, clopidogrel, and ASA plus dipyridamole.

This is the best studied, cheapest, and most widely used antiplatelet medication. In 1994 the Antiplatelet Trialists' Collaboration published a meta-analysis of randomized trials⁴ of prolonged antiplatelet therapy for prevention of stroke (along with various other endpoints). This meta-analysis looked at trials published till 1990, the vast majority of the patients having taken ASA. Amongst 10,000 patients with a past history of stroke or TIA the net event rate for vascular events was estimated at 18% with antiplatelet therapy vs 22% in the controls. (RR 82%, RRR 18%, ARR 4%, NNT 25).

Ticlopidine was the first antiplatelet drug demonstrated to be clearly better than ASA, however it is not commonly used due to a poor side effect profile. The second drug in this class to be released was clopidogrel, along with the results of the CAPRIE trial.

The CAPRIE Trial⁵ was a randomized, double blinded, international trial designed to assess the relative efficacy of clopidogrel (75 mg. once daily) and aspirin (325 mg. once daily), in reducing the risk of a composite outcome cluster of ischemic stroke, myocardial infarction, or vascular death. The population studied comprised subgroups of patients with atherosclerotic vascular disease manifested as, either recent ischemic stroke, recent myocardial infarction, or symptomatic peripheral arterial disease. 19,185 patients were followed for 1 to 3 years. The primary study end point was a composite of ischemic stroke, MI or vascular death. Overall, in all 3 subgroups, there was an annual 5.32% event rate in the clopidogrel group vs 5.83% in the ASA group. (RR 91.3%, RRR 8.7%, ARR 0.51%, NNT 196/per year, p=0.043).

This is not an impressive result and required a huge study to demonstrate it. In the sub group of interest to us, the 6411 patients with a stroke as the qualifying event to enter the trial, clopidogrel did not demonstrate a significant reduction in the composite endpoint. (RRR 7.3% p=0.26). (An argument can be made that this was just type 2 error and we just need a larger study to demonstrate a significant difference - but this was already a huge study). Also remember this data is for a composite end point whilst we are most interested here in secondary stroke prevention. So we can conclude that clopidogrel has not been demonstrated to be superior to ASA in the secondary prevention of stroke. Further analysis of this trial supports the view that clopidogrel is clearly superior to ASA in the prevention of MI in the CAPRIE patient population. (RRR 16.6% NNT 119)⁶

The story of this drug and its combination with ASA is fascinating. By the late 1980's dipyridamole and the combination with ASA was felt to be no better than ASA alone. However, in 1987 ESPS-1⁷, using a combination of ASA 330 mg. plus dipyridamole TID for 2 years in patients with prior ischemic stroke or TIA, showed a RRR of 38% for subsequent stroke. This was far higher than expected from previous studies on ASA alone. There were a lot of questions about the dosing of ASA and the conflicting data of ESPS-1 vs previous trials, so a larger 2x2 factorial design study, ESPS-2⁸ was undertaken in 1989 and published in 1996. Placebo was compared to ASA 25 mg. BID, dipyridamole 200 mg. extended release, and the combination of ASA plus dipyridamole. Patients with a prior TIA or ischemic CVA were enrolled and followed for 2 years in this international multicenter randomized double blinded trial done in Europe.

The results were impressive. 15.78% of the placebo group, 13.21% of the dipyridamole group, 12.93% of the ASA group and 9.95% of the combination group suffered a recurrent stroke. There was no statistical difference between the 2 active treatment groups and with a RRR of 18%, the low dose ASA group was in line with previous trials on ASA. The surprise was the additive effect seen in the combination group. (RR 63%, RRR 37%, ARR 5.83%, p<0.001, NNT 17 over 2 years). There was no difference in the death rate in the 4 groups and no excess of MI in the combination group.

From the evidence it is clear the first choice in secondary prevention should be a combination of low dose ASA plus extended release dipyridamole. It is superior to ASA or clopidogrel alone. It is marketed in Canada under the brand name Aggrenox.

What about the combination of ASA and clopidogrel? There is good evidence to use it for ischemic heart disease¹⁴ but the evidence is lacking for secondary prevention of stroke. This is an area of active research and results of several ongoing trials are eagerly awaited.

Hypertension and the Secondary Prevention of Stroke

Hypertension is a major risk factor for stroke and the treatment of hypertension is a powerful means for primary prevention of stroke. Numerous randomized placebo controlled trials, using primarily diuretics and beta blockers, established this by the 1970's.

Concern about cerebral perfusion in patients with known cerebrovascular disease and especially those with significant carotid disease, meant extrapolation of primary prevention data regarding antihypertensive treatment was resisted till well into the 1990's.

So what is the evidence that treatment of hypertension will reduce the risk of stroke in patients who have already suffered a stroke or TIA? This can be reviewed most efficiently by looking at a recent systematic review of the topic published in the Journal STROKE in November 2003¹⁵. The authors, through an exhaustive search of the world medical literature, identified 7 placebo controlled trials that assessed the effect of lowering blood pressure in patients with prior stoke or TIA. They used meta-analytic techniques to combine these results and came to several not unexpected conclusions. The authors reported the results as odds ratios but these have been converted to relative risks for this article. The outcome, recurrent stroke, occurred in 11.46% of the placebo group, and 8.86% of the treatment group. (RR77%, RRR23%, ARR2.6%, NNT 38, p=0.005). Interestingly these results were heterogeneous depending on the agent used, but please use caution interpreting the results, as the numbers are small in each drug class except diuretics.

Overall diuretics decreased the risk of recurrent stroke, ACE-I reduced the risk of MI and the combination of both drugs reduced both. Four of the seven trials used fixed drug combinations and enrolled both hypertensive and non hypertensive patients, bringing up a lively debate on whether the effects seen are related to the drugs used, or the blood pressure drops seen, in the treatment arms. If the effects seen were due to the blood pressure drop, this brings in to play the question of the definitions of hypertension and target blood pressures.

Atrial Fibrillation and Secondary Prevention

Atrial fibrillation is a potent risk factor for ischemic stroke. This dysrhythmia affects 2-5% of the general population over the age of 60, but is found in 15% of all stroke patients16. The use of anticoagulation with warfarin is now well established as the preferred method for primary prevention of stroke in this patient population, as long as the patient is not considered low risk. (Low risk patients have no history of previous stroke or TIA, no treated or untreated hypertension, no diabetes, and no symptomatic coronary artery disease.)¹⁷ In low risk patients there is no substantive advantage for anticoagulation over antiplatelet agents.

What about secondary prevention? What are we to do for a patient who has already suffered a stroke or a TIA associated with atrial fibrillation?

This question was addressed in 1995 by a Cochrane Review.¹⁶ This review looked at the world literature in a systematic fashion, and performed a meta-analysis on the results. It combined the results of the EAFT from 1993 and VA-SPINAF from 1992. Between the 2 trials 485 patients were included. Anticoagulation reduced the risk of recurrent stroke by nearly 2/3 and the risk of all vascular events over 1/3. (RR 39.5%, RRR 60.5%, p<0.0001, ARR 13.7%, NNT 7 for recurrent stroke).

So the evidence is clear. Patients should be encouraged to stop smoking, treat their lipids and blood pressure to target, take ASA- dipyridamole sustained release for secondary stroke prevention and warfarin if they have atrial fibrillation and a risk factor, or have AF and have already had an event.

- Farokh Buhariwalla

1. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002 Jul 6; 360(9326): 7-22.
   nbsp;
2. Wolf, PA et al, Cigarette Smoking as a Risk Factor for Stroke. The Framingham Study. JAMA.1988 Feb 19;259(7):1025-9
    
3. Gill, JS et al, Cigarette Smoking. A Risk Factor for Hemorrhagic and Nonhemorrhagic Stroke. Arch Intern Med. 1989 Sept;149(9):2053-7
    
4. Collaborative Overview of Randomized Trials of Antiplatelet Therapy-I: Prevention of Death, Myocardial Infarction, and Stroke by Prolongs Antiplatelet Therapy in Various Categories of Patients. Antiplatelet Trialists' Collaboration. BMJ 1994 Jan 8; 308(6921):81-106.
    
5. CAPRIE Steering Committee. A randomized, blinded, Trail of Clopidogrel versus aspirin in patients at risk of ischemic events (CAPRIE), Lancet 1996; 348:1329-39.
    
6. Cannon, PC, Effectiveness of Clopidogrel Versus Aspirin in Preventing Acute Mycocardial Infarction in Patients with Symptomatic Atherothrombosis (CAPRIE Trial). Am J Cardiology 2002 Oct 1;90:760-762
    
7. European Stroke Prevention Study. ESPS Group. Stroke. 1990 Aug; 21(8): 1122-30
    
8. Diener HC et al, European Stroke Prevention Study. 2. Dipyridamole and acetylsalicylic acid in the secondary prevention of stroke. J Neurol Sci. 1996 Nov; 143(1-2): 1-13.
    
9. Yusuf S et al, Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med. 2001 Aug 16; 345(7): 494-502.
    
10. Rashid P, Leonardi-Bee J, Bath P. Blood pressure reduction and secondary prevention of stroke and other vascular events: a systematic review. Stroke. 2003 Nov; 34(11): 2741-8.
     
11. Koudstaal PJ. Anticoagulants for preventing stroke in patients with nonrheumatic atrial fibrillation and a history of stroke or transient ischemic attacks (Cochrane Review). In: The Cochrane Library, Issue 4, 2003. Chichester, UK: John Wiley & Sons, Ltd.
     
12. van Walraven C, et al, A clinical prediction rule to identify patients with Atrial Fibrillation and a low risk for stroke while taking aspirin. Arch Intern Med. 2003 Apr 28; 163(8): 936-43.