Neurology

د. حسين محمد جمعه

اختصاصي الامراض الباطنة
البورد العربي
كلية طب الموصل
2012

Secondary prevention of ischaemic stroke

Key points
After a first stroke the risk of early recurrent stroke and transient ischaemic attacks is high, and rapid specialist assessment is required
This is particularly the case for elderly patients with hypertension, prolonged symptoms, and unilateral weakness or dysphasia
Most patients with atherothromboembolic ischaemic stroke or transient ischaemic attacks should receive antiplatelet therapy (aspirin), antihypertensive treatment (perindopril and indapamide), and cholesterol lowering treatment (simvastatin)
Most patients with cardioembolic ischaemic stroke or transient ischaemic attacks should receive warfarin

Clinical tips

You should refer any patient who has potentially had a stroke or transient ischaemic attack as soon as possible
You should start secondary prevention as soon as possible because the risk of recurrent stroke is highest in the first few weeks.
You should start treatment to lower the patient’s blood pressure and cholesterol, even where these are within the "normal" range.


What is ischaemic stroke?
Ischaemic stroke is the result of an interruption of the blood supply to a volume of the brain and accounts for around 80% of all strokes. The remaining strokes are haemorrhagic stroke (15%) and
subarachnoid haemorrhage (5%). Haemorrhagic stroke may be the result of a structural abnormality such as an arteriovenous malformation. We will not cover haemorrhagic stroke or subarachnoid haemorrhage in any more detail in this module.

Ischaemic stroke may be caused by embolism of thrombus formed elsewhere in the circulation (for instance from the heart, aortic arch, or proximal internal carotid artery) or by in situ occlusion, usually of small perforating lenticulostriate arteries.
Clinically, strokes may be classified according to the clinical features and the brain region and vascular territory presumed to be involved; that is, the deep white matter (lacunar syndromes) and the posterior or anterior circulation.

Strokes of the anterior circulation are subdivided into those that involve all functions in the affected hemisphere (total anterior circulation syndrome) and those that involve only some of these functions (partial anterior circulation syndrome).
This clinical classification does not imply any aetiology, but it does provide a useful prediction of outcome; with patients who have total anterior circulation syndrome doing poorly and those with lacunar syndromes doing well.

Who gets it?

The incidence of ischaemic stroke increases with age. Stroke is more common in patients with hypertension, diabetes, atrial fibrillation, valvular heart disease, carotid or aortic arch atheroma, those with a family history of stroke, and cigarette smokers.
In younger patients, other risk factors are important, including thrombophilic tendencies, vasculitides, carotid or vertebral artery dissection, and patent foramen ovale.

How do I diagnose it?

Ischaemic stroke is characterised by the sudden onset of focal neurological dysfunction as a result of focal brain ischaemia that lasts more than 24 hours and cannot be better explained by other neurological disease. Important differential diagnoses include haemorrhagic stroke, migraine, epilepsy with Todd's paresis, venous sinus thrombosis, and hypoglycaemia.

Although any neurological function may be affected, the most common presentation is with weakness of the face, arm, or leg (or their combination), with or without speech disturbance or hemianopia.
Headaches and seizures are uncommon at the onset of stroke, occurring in only 5% and 2% of patients respectively.
Previously, much was made of the distinction between transient ischaemic attacks (defined as stroke, but with symptoms lasting less than 24 hours) and stroke.


It is now clear, however, that as far as secondary prevention strategies are concerned, the risks of recurrence and the benefits of early intervention are broadly similar. Indeed, increasing evidence from imaging suggests that irreversible brain injury may occur after even brief transient ischaemic attacks. For these reasons, the secondary prevention strategies described below for ischaemic stroke should be pursued with equal vigour in patients with transient ischaemic attacks.

Clinically, ischaemic and haemorrhagic strokes cannot be distinguished reliably,

although coma or severe headache at the time of presentation supports a diagnosis of haemorrhagic stroke.
The most robust discrimination comes from early computed tomography, but as time passes, computed tomography becomes unreliable and magnetic resonance imaging is required.
Given the relative availability of computed tomography and magnetic resonance imaging, early referral with prompt imaging is central to the rational organisation of services.

How do I treat it?

The goals of treatment in acute stroke are to:
Reduce any resulting disability by strategies to Reduce brain injury
Promote recovery and
prevent complications
Prevent recurrent stroke.

For a decade, thrombolysis within three hours of the onset of symptoms has been known to be a highly effective treatment for ischaemic stroke.
Recently, this time window has been extended to four and a half hours. Despite this, thrombolysis for patients with stroke in the United Kingdom is still not available in all stroke centres; extension of stroke services in order to bring the benefits of thrombolysis to all who might benefit is a matter of some priority.

Aspirin given within the first 48 hours has about the same impact. This is because, although the benefit achieved is substantially less (about 10 fold less) than that seen with thrombolysis, many more patients are able to take aspirin.
Finally, rehabilitation in the setting of a specialist stroke unit rather than a general medical ward has a substantial and prolonged favourable effect on outcome.
For the doctor in primary care, the task is firstly to recognise potential stroke and transient ischaemic attacks, and secondly to have the patient assessed in secondary care as soon as possible.


The difficulty in distinguishing between haemorrhagic and ischaemic strokes on clinical grounds (without computed tomography of the brain), and the potential benefits (in selected patients) of early intervention, such as carotid endarterectomy, means that management of brain attack exclusively in primary care has no place, and referral to secondary care is advised in all patients with a stroke, provided that they want to attend.

How do I prevent it happening again?

Because the risk of recurrent stroke after transient ischaemic attack or minor stroke is highest in the first few weeks, the sooner secondary prevention strategies are introduced, the better the chance of reducing that risk. Where there is a delay prior to specialist review (and many centres now aspire to run a same day or next day service), the opportunities for stroke prevention are lost.

The key is to predict which patients are at highest risk of early recurrence. One model (the ABCD2 system) uses:
A - age (≥60 years) (1 point)
B - blood pressure at presentation (systolic ≥140 mm Hg or diastolic ≥90 mm Hg) (1 point)
C - clinical presentation
Unilateral weakness (2 points)
Speech impairment without weakness (1 point)
Neither speech impairment or weakness (0 points)
D - duration of symptoms
≥60 minutes (2 points)
10-59 minutes (1 point)
<10 minutes (0 points)
D - presence of diabetes (1 point).
When added together, the total ABCD2 score ranges from 0 (low risk) to a maximum score of 7 points (high risk).

The risk of stroke within two days of the initial event has been shown to be :

1% in patients with a score of less than 4
4.1% in patients with a score of 4 or 5
8.1% in patients with a score of 6 or 7.
You should start patients who have had a suspected transient ischaemic attack and who are at high risk of stroke (that is, with an ABCD2 score of 4 or more) immediately on aspirin and a statin, such as simvastatin, and refer them for urgent specialist advice.


What treatments should I use to prevent recurrence?
Treatments to prevent recurrence of stroke can be divided into:
Treatments that are appropriate for most patients (global interventions)
Specific interventions for particular groups of patients.
These treatments have all been tested in patients with ischaemic stroke and do not apply to haemorrhagic stroke or subarachnoid haemorrhage.

For patients with an underlying medical condition that is known to increase the risk of stroke, the management of that condition should be optimised. For example, patients with diabetes and those with vasculitis, such as temporal arteritis, should be managed appropriately.

In patients with underlying conditions known to increase the risk of stroke for which no specific treatment exists (for example, those on immunosuppressive regimens after renal transplantation), it seems reasonable to take a more aggressive approach to the non-specific secondary prevention measures outlined below.
Although these measures may be initiated in secondary care, responsibility for monitoring and, where necessary, intensifying them often falls to the primary care doctor.
Guidelines for these treatments are taken from the national clinical guidelines for stroke

What global interventions can I use?

These global interventions are recommended for all patients unless there is a good reason not to use them. Specifically, treatments to lower blood pressure and blood cholesterol levels are indicated, even where the values fall within the "normal" range. General practitioners generally manage these interventions

Antiplatelet treatments

The beneficial effects of antiplatelet treatments are well established; over two and a half years the risk of serious vascular events is decreased from 21.4% to 17.8% (number needed to treat, 28) in patients with a prior stroke or transient ischaemic attack.

The first line antiplatelet treatment is aspirin. In one systematic review, no significant difference was seen between patients who received low doses of aspirin (<75 mg/day) and those who received higher doses of aspirin (≥75 mg/day), and it seems reasonable to use a dose of 75 to 150 mg/day.

The major risks are of intracerebral haemorrhage (number needed to harm, about 300 in patients with a prior stroke or transient ischaemic attack) and severe gastrointestinal haemorrhage (number needed to harm, about 270 in patients at high vascular risk).
A recent study suggests that, in patients with gastrointestinal disturbance, the introduction of a proton pump inhibitor in addition to aspirin is associated with a lower risk of bleeding than switching the patient to an alternative antiplatelet agent, such as clopidogrel.


More recently, the efficacy of alternative antiplatelet agents has been tested and monotherapy with dipyridamole seems to be no more effective than aspirin. A throbbing headache is a common side effect of dipyridamole, so you should use it with caution in patients with migraine.
In patients at vascular risk, including those with previous stroke or transient ischaemic attack, the combination of aspirin and dipyridamole reduced vascular events over about two years from 15.8% to 14.2% (number needed to treat, 62).

Monotherapy with a thienopyridine (clopidogrel or ticlopidine) seems to be marginally more effective than aspirin alone over around two years (18.3% to 16.8%; number needed to treat, 68). Ticlopidine has been found to lead to severe neutropenia in around 1% of patients (number needed to harm, 118), and clopidogrel is associated with a mild excess of skin rashes and diarrhoea.

Clopidogrel plus aspirin is no more effective than either clopidogrel or aspirin alone.

Clopidogrel is also not superior to aspirin plus dipyridamole. The risk of bleeding in patients who receive aspirin plus clopidogrel seems to be increased over that seen in patients who receive clopidogrel alone.

Blood pressure lowering

Even patients whose blood pressure falls within the "normal" range seem to gain a substantial benefit from a reduction in blood pressure.
The PROGRESS trial tested perindopril (4 mg) with or without indapamide (2.5 mg, at the discretion of the treating doctor) in patients with a previous transient ischaemic attack or minor stroke. Over four years, major vascular events were reduced from 19.8% to 15.0% (number needed to treat, 21), and this benefit was seen at all levels of blood pressure. The benefits were even more marked in those who received indapamide in addition to perindopril (13% versus 20.7%; number needed to treat, 13).

No target has been established for how low patients' blood pressure should be taken. In the first instance, GPs should try to gradually increase the doses of perindopril with or without indapamide to the recommended doses (4 mg and 2.5 mg respectively). If side effects become troublesome, it is reasonable to maintain patients on the highest doses that can be tolerated without side effects developing.

Some concern exists that decreasing blood pressure during the acute phase of stroke may reduce cerebral perfusion; the PROGRESS trial started treatment after patients had been clinically stable for two weeks.
Given the magnitude of the benefit of treatment and the risks of it not being initiated, however, it is reasonable to start treatment immediately in those patients who present with transient ischaemic attacks and, in those with stroke, when the patient has been stable for a week (or at hospital discharge, if earlier).

Cholesterol lowering

Even in patients whose levels of cholesterol lie within the "normal" range, reduction in cholesterol over five years is beneficial.
Simvastatin (40 mg at night) has been shown to reduce the risk of major vascular events from 29.8% to 24.7% over five years (number needed to treat, 20), but has no apparent effect on the risk of recurrent stroke. Even in patients without a prior history of coronary heart disease, atorvastatin (80 mg per day) has been shown to reduce the risk of recurrent stroke from 13.1% to 11.2% over about five years (number needed to treat, about 50).


The maximum dose of rosuvastatin that you can give to patients of South Asian origin is 20 mg daily.
The benefits of treatment are probably the generic effects of HMG-CoA reductase inhibition. It is likely that the benefits of statins are a class effect; the magnitude of which relates to the extent to which levels of low density lipoprotein (LDL) cholesterol are reduced.

There is also a small amount of laboratory evidence suggesting that statins may have additional neuroprotective properties (through inhibition of rho kinases), and this provides a further impetus for the prompt initiation of statin therapy following a stroke.

Diet

There is no evidence to support a specific role for dietary modification after a stroke, but it is reasonable to advise patients to adopt a healthy diet if they do not already do so.
You need to be aware of important dietary interactions with different drugs used in the secondary prevention of stroke.

For example:

Patients on simvastatin should avoid grapefruit juice
This is because grapefruit juice has been shown to increase the concentration of some CYP3A4 substrates such as simvastatin; even modest quantities of grapefruit juice can lead to a large increase in the concentration of simvastatin, thereby increasing the risk of rhabdomyolysis
Patients on warfarin should avoid cranberry juice
Cranberry juice can cause the international normalised ratio to rise through an unknown pharmacodynamic interaction and so may cause bleeding in patients on warfarin.

Exercise

Cardiorespiratory training after a stroke leads to improved functional mobility and maximum walking speed, so it is important for patients with residual impairment after their initial stroke. It is reasonable to suppose that general encouragement to increase the amount of exercise taken is unlikely to cause harm and may be beneficial in all patients in terms of the risk of recurrent stroke and their general health.

Smoking cessation

There is no evidence to support a specific role for smoking cessation after a stroke, but the causal links between cigarette smoking and the risk of a first stroke were established by Richard Doll more than 50 years ago. Given the many health benefits from smoking cessation, including the benefits to vascular health, this should be strongly encouraged. Manufacturers of nicotine replacement therapies advise against their use in patients with a history of stroke, but any potential harm (for which little evidence exists) needs to be set against the potential benefits of smoking cessation.


What specific interventions can I use?
Warfarin for atrial fibrillation
In patients with valvular and non-valvular atrial fibrillation, the benefits of warfarin therapy at a target international normalised ratio (INR) of 2.0 to 3.0, compared with placebo and aspirin, are well established.
Recent interest has focused on whether combination treatment with low doses of warfarin plus antiplatelet therapy might be as effective as full doses of warfarin.

Regimens that involve fixed low doses of warfarin (which give an INR <1.4) have not been successful. Recent data suggest that both patients with valvular atrial fibrillation and patients with non-valvular atrial fibrillation who also have a history of embolic events benefit from the combination of an antiplatelet agent with warfarin (target INR 1.4 to 2.4), compared with anticoagulation alone. This finding from one small randomised controlled trial needs to be confirmed in other studies.

Some concerns exist that the quality and monitoring of prescriptions for warfarin may be substantially lower in the real world than in clinical trials, which could lead to a reduced efficacy and an increased risk of haemorrhagic complications.
This provides a powerful impetus to improve the community based use of warfarin.

When should warfarin be started?

There is no evidence from clinical trials to guide the timing of starting anticoagulation, but it seems reasonable to start it as soon as it is safe to do so. In patients with transient ischaemic attacks and atrial fibrillation, warfarin might be started once a computed tomography image of the brain has excluded haemorrhage.

For patients with a minor stroke, warfarin may be started once they have been clinically stable for 48 to 72 hours.
For those with a more major stroke, the substantial risk of haemorrhagic transformation peaks within the first week, and warfarin may be started after two weeks.
Whether or not fixed doses of oral thrombin inhibitors can be substituted for warfarin is currently being studied.

Carotid endarterectomy for significant ipsilateral stenosis

Carotid endarterectomy is indicated in patients with significant (>50%) stenosis of the carotid artery on the side of the stroke.
A patient may have significant carotid stenosis despite the fact that you cannot hear an audible bruit.


The risks of surgery are substantial, however, and the benefits of the procedure are most marked in men, people older than 75 years, and when the operation is performed within two weeks of the stroke. This provides further impetus for the rapid assessment of such patients.
Ongoing studies will help to determine whether endovascular treatment of carotid stenosis is more effective or less hazardous than open endarterectomy.

Clinical situations where the optimal strategy for secondary prevention is not known

In some situations, the optimal strategy for secondary prevention is not known. For example:
Aortic arch atheroma - arch atheroma conveys a substantial increase in the risk of recurrent stroke, and the optimal management of this is the subject of the ARCH study
Lupus anticoagulant or antiphospholipid syndrome - lupus anticoagulant may be found particularly in younger women with a stroke who perhaps have a history of deep vein thrombosis, pulmonary embolism, or miscarriage.

Although it does seem to increase the risk of stroke, optimal management is not known

Patent foramen ovale - although common in the general population, patent foramen ovale, particularly when associated with an atrial septal aneurysm, seems to increase the risk of recurrent stroke. Whether or not closure of patent foramen ovale reduces this risk is being studied in a clinical trial.

Increased levels of homocysteine - although increased levels of homocysteine seem to convey an increased risk of stroke, interventions that decrease the levels of homocysteine do not reduce the risk of recurrent stroke.
Strokes associated with intracranial arterial stenosis - although these patients are at increased risk of recurrent stroke, the warfarin versus aspirin for symptomatic intracranial disease (WASID) study showed that treatment with warfarin was substantially more hazardous than aspirin in these patients.

When should I refer my patient?

A clear difference exists between what might be identified as "best practice" and what is practicable in the NHS at the current time. Ideally, all patients with symptoms and signs of acute stroke should call 999 and attend an emergency department immediately. This will allow their stroke to be managed by the brain attack team in a safe clinical environment, with access to the diagnostic tools and treatments that will maximise their chance of a good outcome.

Many emergency departments, however, do not have the resources to support such a service, and many hospitals do not have an acute stroke or brain attack team. In these circumstances, local guidelines agreed between general practitioners, emergency departments, stroke doctors, and neurologists should inform what it is possible to achieve locally.

The case is similar for patients with transient symptoms. Ideally, they should all be assessed by a stroke doctor or neurologist on an emergency basis, and certainly within 24 hours. However, in the United Kingdom at least, this is not possible given current resources.


Some patients, however, do need to be seen immediately and should be triaged through emergency departments. These include patients on warfarin and those who fulfil the criteria outlined above for identifying people at high risk of early recurrence (scores of 6 or 7 on the ABCD2 scale).

Patients on warfarin should be assessed immediately as they are at substantial risk of their symptoms being the result of intracerebral haematoma.
All other patients should be seen as soon as is practicable - and current guidelines suggest within one week at most - by the local neurovascular team.