This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

CAMHS won't see you now

More commissioning risk for GPs

The definition is of focal neurological symptoms or signs, of presumed vascular origin that fully resolve in 24 hours. When confronted with a patient you think may have had a TIA it is worth asking four questions:

·Are the neurological symptoms focal?

·Are they negative rather than positive? Negative means loss of function (eg paralysis, numbness, loss of vision), whereas positive symptoms include tingling, parasthaesiae, twitching and distortion of vision.

·Was the onset sudden?

·Were the symptoms maximal in onset rather than progressing over a period?

If the answer to all four questions is yes, the symptoms are almost certainly caused by vascular pathology. If even one of the answers is no, cerebrovascular disease is much less likely and alternative diagnoses should be considered.

There is no way to accurately distinguish between haemorrhage and infarction without brain imaging. Certainly headache, neck stiffness and clouding of consciousness are more often seen in haemorrhage but can also occur with infarction.

So unless making a diagnosis is of no relevance to the patient, and about the only circumstance under which this would be the case is if the decision has been made to give only palliative care, the patient should have a brain scan.

If a patient refuses admission, as long as they understand it may put their recovery in jeopardy, you have little choice other than to organise a brain scan as an outpatient. Older patients are less likely to get an urgent scan, although these days more than 90 per cent do get a scan eventually.

There is no justification for what can only be discriminatory behaviour. Please keep up the pressure on acute services to deliver high-quality and equitable services. Your QOF targets don't have an age cut-off!

Aspirin remains the antiplatelet drug of choice for the vast majority of patients who have had an ischaemic stroke. It should be given as soon as the diagnosis has been confirmed by brain imaging at a dose of 300mg, reducing to 75mg after two weeks.

The recently published NICE Health Technology Assessment recommends that all patients should also be given dipyridamole MR 200mg bd for the first two years after the stroke. However, not all stroke physicians are convinced that this is really necessary for everyone and the National Clinical Guidelines for Stroke leaves the choice of anti-thrombotic treatment more flexible.

Bleeding into an infarct is common (haemorrhagic transformation). In the majority of patients this does cause any noticeable deterioration in symptoms and there are probably many people who have some haemorrhage that is never detected because repeat brain scanning is not routine.

Where it is identified I usually stop the aspirin for a couple of weeks and then restart it. The benefits of preventing further infarction far outweigh any theoretical risk of precipitating further bleeding. The same is true for patients who have had a previous primary intracerebral haemorrhage who then have a cerebral infarction.

When aspirin is being used following stroke, treatment should be continued indefinitely, unless the patient develops side-effects or the drug proves ineffective. Obviously it should be stopped if there is significant aspirin allergy.

Dyspepsia is unusual when low-dose aspirin is used (75mg) but when it does occur I usually continue the drug, but add in a proton pump inhibitor.

If helicobacter positive peptic ulceration develops I stop the aspirin, give eradication therapy and then restart the aspirin again with gastro-protection.

Clopidogrel also carries the risk of precipitating gastric bleeding and so there is little point in switching treatment in the hope that side-effects will be reduced. I do not favour the use of dipyridamole as a single agent as I am unconvinced of its efficacy.

So-called 'aspirin failures' (patients who have further events while on the drug) are difficult as there is no evidence base for treatment. Despite in-vitro evidence suggesting the possibility of aspirin resistance there have been no studies to show that increasing the dose of aspirin is effective. My own practice is to add dipyridamole MR to the aspirin and if attacks continue switch the patient over to single agent clopidogrel.

Warfarin is more effective than antiplatelet drugs in preventing cardioembolic stroke. So patients in atrial fibrillation, those with mitral valve disease, and those who have had a stroke as a result of a mural thrombus following myocardial infarction should be anticoagulated.

However, warfarin obviously also causes a significantly increased risk of bleeding. Therefore there is always a decision to be made balancing risk and benefit.

There is no formula that can be applied and it is often a difficult judgment to make. Clearly if the patient is very frail and at risk of falls then the risk of subdural haematoma is probably greater than the risk of stroke. If the patient has significant cognitive difficulties, or is considered unlikely to be able to comply for other reasons with the treatment, it is better to stick to aspirin as the alternative.

Old age alone is not a reason for avoiding warfarin. In fact the benefits are greater because the risk of stroke is higher. As a profession we do not anticoagulate patients nearly often enough. The 2004 National Sentinel Audit of Stroke showed that less than half of those appropriate for treatment received it.

TIA should be treated as a medical emergency in the same way as you would treat someone who presented with crescendo angina. The risk of stroke in a hypertensive patient with a hemispheric TIA is about 20 per cent in the first month, with most strokes occurring in the first week.

Unless you are suspicious that the TIA was due to a haemorrhage, and that is unusual, start aspirin 300mg immediately. Continue with any existing antihypertensive medication, but unless the blood pressure is above 220/120 don't initiate new blood pressure-lowering treatment as this could exacerbate the neurovascular disease, particularly if there is tight carotid stenosis.

Refer to your neurovascular clinic with the expectation that they will be seen and investigated within the week. If your local service cannot offer such a service then make a fuss because your patients are being put at unnecessary risk.

If a patient has more than one TIA in a week (crescendo TIA) the National Guidelines recommend admission.

Stroke as a disease has been slow to be acknowledged as a treatable disease that needs urgent attention. There are now many hospitals that can offer appointments in neurovascular clinics within seven to 14 days but even this is not good enough as there will be patients who have a stroke before being seen.

The Department of Health in England is now taking stroke seriously and I expect there to be policy announcements that should improve the situation shortly. The National Audit Office is preparing a report on stroke services to be published at the end of this month that will highlight areas that need reform. I would be very surprised if this wasn't addressed.

Cocaine can cause stroke by inducing a vasculitis that leads to cerebral ischaemia. There are no studies I have been able to identify that describe the natural history of cocaine-induced stroke, but I would expect that when the drug is stopped the inflammation will settle and the risk of stroke returns to normal. Amphetamine-based drugs and ecstasy can also cause stroke, usually haemorrhage, probably as a result of causing acute hypertension.

The evidence points to HRT increasing the risk of cerebrovascular disease and it should therefore not be used if that is the main rationale. Clearly HRT is effective at reducing menopausal symptoms and osteoporosis and so far there has not been a study published that looks at overall mortality and morbidity when HRT is used for long periods.

However, my advice is that HRT should be stopped after a stroke and women who have not yet had any vascular pathology should not take HRT for more than a few years after the menopause unless there are very specific reasons for doing so.

Very elderly people were not included in many of the trials looking at risk factor reduction so the evidence for treating this age group is not as strong as for younger patients. But that is no reason to suppose drugs that work on younger patients would have no effect on someone over the age of 75.

Indeed logic would suggest that as these patients are at greater risk of stroke the benefit might be greater. The National Stroke Audit from 2004 shows older people are often undertreated and are much less likely to have been given advice about changing lifestyle than their younger counterparts (see graph below).

If a patient (of whatever age) is very frail with multiple co-morbidities a more pragmatic approach is clearly sensible. For example if you predict life expectancy of less than two years there is probably little point in starting them on a statin, as the trials really only show a major separation in mortality compared with placebo over a longer period.

All patients should be managed on an individual basis ­ but don't allow ageism to

affect your judgment.

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say