Skip to content

Focus on stroke: Treating stroke

14 May, 2012

Stroke is a desperate, time-critical emergency. While it’s happening (the ‘acute’ phase), nearly 2 million brain cells are dying every minute – disabling somebody in seconds. It doesn’t hurt, so people may not be aware they’re having a stroke until they suddenly realise they can’t speak, move, walk, talk or swallow. Penny Bailey looks at how stroke used to be treated and what has changed for patients today.

Stroke screams emergency, but until a few years ago doctors were powerless to do anything about it. People who turned up at hospital suffering an acute stroke were either sent home with an aspirin or admitted to a general ward where they were looked after by general doctors following their own individual protocols. There were no stroke specialists, research was sporadic and bitty, and there was no evidence base to inform treatment. For patients that meant death or severe disability were all too likely.

The hopelessness that paralysed stroke medicine had its roots partly in the complexity of the condition, and partly in the fact we were (and are still) very restricted in how far we can see into the brain. We’ve yet to precisely map the labyrinthine pathways and intricate cellular processes of a healthy brain, let alone fully understand how stroke destroys it.

Medicine is rarely just about biology, of course, and the problem of stroke has been compounded by a failure of governments and health services to prioritise and dedicate resources to it, by low levels of public awareness of what stroke is, and (in the UK) by anxiety about changes to the NHS.

Credit: Annie Cavanagh. Wellcome Images

Clot busters

The 1990s were the decade that revolutionised stroke medicine with two breakthroughs that gave people who had suffered a stroke the first glimmer of hope.

One was the arrival of the clot-busting drug tissue plasminogen activator (tPA)1 in 1995 for patients with stroke caused by a blood clot in the brain (ischaemic stroke). Clot busters like tPA restore the blood supply to oxygen-starved brain cells, which recover rapidly – sometimes to spectacular effect. According to stroke consultant Paul Guyler [PDF] at Southend University Hospital, people who have been unable to speak or walk suddenly find they are restored to their former selves, in what clinicians gleefully call the ‘Lazarus effect’.

There is a caveat: if too much brain tissue has died, restoring the blood supply risks the second major type of stroke, haemorrhagic stroke (bleeding into the brain tissue).

To minimise that risk, clinicians only give tPA up to 4.5 hours after the beginning of the stroke. Identifying precisely when that was is often difficult because stroke is extremely stealthy. One in five happens in the night; people wake up in the morning with symptoms but no idea when or how they got them (a ‘wake-up stroke’). Even if they are awake when it strikes, stroke can obliterate both function – and the memory of it – so completely that people aren’t even aware it’s been lost. Often they are the last to know they’ve had a stroke; it takes someone else to notice they are slurring or lopsided.

Even if someone does realise what’s happening, they still have to get to a hospital with a stroke consultant, get prioritised in A&E, then assessed and scanned, all within 4.5 hours. And even then not every ischaemic stroke patient will be given tPA. It’s a powerful drug that will break down clots anywhere in the body, making it unsuitable for someone with a stomach ulcer or who has recently had surgery. And for people over the age of 80 (who are likely to have fragile blood vessels), or who are having a very mild, self-limiting stroke (known as a ‘transient ischaemic attack’ or TIA), the risk of harm from a haemorrhage outweighs any potential benefit from tPA. That meant that although ischaemic stroke is the major type of stroke, accounting for 80 to 90 per cent of all cases, only a small percentage of people with an ischaemic stroke could receive tPA.

Credit: Justine Desmond. Wellcome Images

Stroke units

Nevertheless, here finally was something that could check stroke in its pitiless stride – and tPA had unexpected knock-on effects far beyond its efficacy as a drug. One was the fact that stroke was finally designated the medical emergency it so devastatingly is – and treatment moved from the home to hospitals.

Another was its synergistic effect on the second major breakthrough of the 1990s: the proliferation of specialist stroke units. The first of these had been established in the UK a decade earlier but by the arrival of tPA in 1995 there were still only three of them. Clot-busting treatments required rapid imaging facilities and specialist clinicians, prompting more hospitals to reconfigure their services and set up dedicated stroke units – a development that was given impetus by evidence that began to build up throughout the 1990s supporting their effectiveness.

In the acute phase of a stroke, staff on a stroke unit can do a lot of damage limitation. They can treat infections (stroke patients are susceptible to pneumonia because their lung function may be suppressed), maintain oxygen levels, and control cholesterol and blood pressure. Crucially, they can prevent a second stroke – which is always a high risk after the first and usually far more severe, often fatal, because the brain is already weakened and vulnerable. According to Professor Keith Muir at the University of Glasgow, while there is still no formal evidence that acute stroke unit care is indeed more effective than other types of hospital care, individual studies such as Professor Bent Indredavik’s 1991 Trondheim study 2 strongly indicate that it is.

The evidence that stroke units improve long-term outcomes once the acute stage is over is more robust. Specialist stroke nurses, physiotherapists and speech therapists can help kick-start the recovery process, by getting patients up, moving around and talking as soon as possible. After a stroke the brain abandons the dead bit of tissue and tries to reconnect all the neurons around it to restore as much function as it can. But it can only do that if the patient is actually trying to retrieve that lost function and learning to walk and/or talk again in the same way a baby does – by doing it, over and over again.

The 1991 Nottingham Stroke Unit study was one of the first UK studies to compare the results of a dedicated stroke rehabilitation unit with those of a general hospital ward. Key evidence for the effectiveness of specialist stroke-unit care was synthesised in 1993 by Professor Peter Langhorne and colleagues3. That meta-analysis, supported by subsequent studies, showed that stroke units improve outcome equally for patients suffering from any type of stroke – ischaemic or haemorrhagic – reducing the odds of death by almost 20 per cent and of death and dependency by 30 per cent.

Even where the evidence for new treatments is systematic and sound, getting them established as standard practice is rarely plain sailing. In the UK the uptake of tPA in the clinic was fraught by concerns about pouring resources into something that only benefited a minority of stroke patients. And a national roll-out of stroke units was hampered by a lack of clear guidelines and mandates at government level, an obsolete hospital infrastructure and public unease about the closure of local general hospitals in favour of specialist centres. And purely logistically, it takes time and an awful lot of management to reconfigure hospital services.

But things are moving in the right direction. According to the Stroke Association, the number of units in the UK has soared from three in 1995 to 150 today. And the National Sentinel Stroke Audits of 2005 and 2010 show the number of patients receiving tPA in the UK grew from 0.1 per cent in 2005 [PDF] to 5 per cent in 2010 [PDF]. There’s still room for improvement – that 5 per cent represented only a third of patients eligible for tPA (they arrived at hospital within 3 hours, were under 80 years and had ischaemic stroke diagnosed by a brain scan). And the Stroke Association says that many hospitals throughout the country still fail to meet quality markers set by the National Stroke Strategy (and NICE guidelines), especially at weekends and in the evening.

Credit: Justine Desmond. Wellcome Images

The devil and the deep blue sea

The uptake of tPA and stroke units may have been frustratingly slow, but on the other hand, new ideas and developments in stroke are proliferating so fast in the clinic that research hasn’t had a chance to catch up and provide a robust evidence base to support their use. Keen to improve the outcome for individual patients, clinicians use interventions that seem obvious intuitively, but for which there is no conclusive epidemiological evidence, on an ad hoc – even routine – basis.

However by raising stroke’s profile – and proving it can be treated in the acute phase, sometimes to dramatic effect – tPA and stroke units have attracted the interest of researchers across the globe, and funds to support them. Trials are ongoing to assess a range of new (and old) interventions – and the field may well look very different in a few years time.

One area dogged by lack of evidence is the interplay between haemorrhagic and ischaemic stroke, which still presents a knife-edge for clinicians. How do you decide whether to treat an ischaemic stroke with blood thinners and risk a major bleed? Or whether to thin the blood in someone with evidence of both ischaemic stroke and bleeding in the brain?

These dilemmas are exacerbated by the fact that although the two types of stroke have different causes and pathways (aetiologies), they share many of the same risk factors. High blood pressure and cholesterol both damage blood vessels in the brain, and diseased small vessels can both bleed and block, so people at risk of hameorrhage are often also at risk of ischaemic stroke.

People are routinely prescribed blood thinners like warfarin or aspirin to prevent an ischaemic stroke (particularly if they’ve already had one). But if their blood vessels are already fragile and they are having small, relatively harmless bleeds in the brain (microbleeds), might anticoagulants like warfarin prevent those microbleeds from sealing themselves off, leading to a severe, large bleed? As yet, we don’t know the answer, but Dr David Werring at UCL is currently leading a five-year research programme that should provide one.

In an attempt to get round the haemorrhage-ischaemia dilemma consultants have also been attempting to surgically extract particularly large, intransigent clots (‘clot retrieval’), using tiny devices, 3-4 mm in diameter. These devices are fed up through a patient’s artery to the brain, on the end of a long wire, to try to pull the clot out or break it up physically.

There’s impressive circumstantial evidence to support the use of these devices: proof-of-principle studies show they do break clots up and clinicians have had very good results using them in individual patients; the technique also makes sense intuitively. As a result, it has fallen into use haphazardly around the world – routinely in Germany and the USA, and on an ad hoc basis in the UK.

But it does carry risks – of damaging the blood vessel wall, and of timing. It’s a fiddly process, which takes an hour or two in the most expert hands. As yet there have been no large-scale studies supplying concrete evidence that they improve patient outcome – something the PISTE trial led by Professor Keith Muir hopes to address.

Credit: Wellcome Library, London. Wellcome Images


Standard imaging techniques in the brain, such as perfusion (which measures blood flow) and diffusion (which measures water diffusion around tissues and hence their integrity) can help to tell clinicians what bits of the brain are alive or dead. But what clinicians really need to know is not what’s dead (which they can’t do much about), but what tissue in a stroke-besieged brain is ailing and going to die without treatment. That’s the tissue that tPA can save, aptly called the penumbra (half shadow).

Dr Stephen Payne and colleagues at Oxford are looking at new imaging techniques to measure how ‘well’ or viable brain cells are. Specifically, they are focusing on pH and oxygen consumption, both of which can be inferred from imaging. Lower pH (indicated by slower chemical reactions) and reduced oxygen consumption can suggest that cells are struggling but still alive and salvageable. If those techniques turn out to be reliable, it means clinicians won’t have to worry so much about pinpointing when a stroke began. They’ll be able to throw the 4.5-hour rule of thumb away and decide whether to give tPA based on what’s happening in an individual patient’s brain.

Payne is working closely with the newly opened stroke unit in Oxford, highlighting another important role of centralised stroke units: to facilitate research as well as to give patients the best possible care.

The WAKE-UP trial is another attempt to get round the 4.5-hour tPA ‘deadline’. A research consortium across six European countries is developing new MRI-based diagnostic techniques for people with wake-up stroke. They hope to be able to identify more precisely when the stroke began – and who could still benefit from clot-busting treatment.

A novel imaging technique developed at the University of Manchester by a team led by Professor Hugh McCann may help with monitoring patients in the critical 48 to 72 hours after the first stroke, when further episodes typically escalate the condition. The fEITER (functional electrical impedance tomography by evoked response) system measures electrical signals deep within the brain – including one known as the REG (rheoencephalography) signal, which varies with changes in blood flow in the brain. McCann and colleagues at Manchester Royal Infirmary have already used fEITER in clinical trials with 40 human subjects to study the brain processes involved in the induction of anaesthesia.

His team is now doing a small pilot study at Hope Hospital, Salford, to determine what mechanism gives rise to the REG signal: whether it is a change in intracranial pressure or a change in cerebral blood flow. Once they have established that, they hope to do a larger trial, to find out whether it can be used on someone who’s just had a stroke to measure what’s happening in the brain continuously over several days. At present, it’s not always possible to know when an unconscious patient is suffering further strokes because people can’t be kept permanently in a large CT or MRI scanner. It may ultimately be possible to refine the system to include automated analysis of fEITER measurements that could automatically raise an alarm.

McCann’s fEITER imaging system may also be suitable for ambulance-based diagnosis and monitoring. It’s lightweight and portable and, once the team have established whether the REG signal is related to changes in blood flow or in pressure within the brain, they may be able to use the signal to detect whether the stroke has been caused by a clot or haemorrhage.


The ‘Lazarus effect’ of tPA has yet to be replicated in haemorrhagic stroke. Haemorrhage accounts for only 10 to 20 per cent of strokes but 50 per cent of those affected die within a month, compared with 10 to 20 per cent of ischaemic stroke patients. Attempts to find a clot-buster equivalent that will arrest and reverse its devastating course have resulted in what Dr Malcolm Macleod at Edinburgh calls a “graveyard” for drugs. Many looked promising in early trials yet for one reason or another failed to fulfil that promise in larger studies.

Recombinant factor VIIa – a clotting factor first used by the Israeli army to prevent bleeding and trauma – reduced haemorrhage volume but without reducing the death rate. Steroids proved to be of no benefit in reducing pressure. The osmotic diuretic Mannitol, which researchers hoped would suck blood out of the brain, had little effect. NXY-059, a powerful scavenger of free radicals developed by AstraZeneca, was eagerly anticipated as a neuroprotective agent but proved disappointing. And glycine blockers aimed at reducing toxic concentrations of glutamate (an amino acid that builds up in the brain during a stroke) also failed to improve patient outcomes.

Credit: Wellcome Library, London. Wellcome Images

The hunt goes on

In search of agents to stem the bleeding, researchers are resurrecting some older clotting drugs like tranexamic acid (an anti-fibrinolytic used to prevent blood loss in surgery) in clinical trials. And a team in Amsterdam is investigating the effects of giving extra clotting factors from platelet donations (the PATCH trial).

One promising avenue was suggested by a pilot study which indicated that aggressively lowering blood pressure reduces the mass of blood (haematoma) that collects in the brain during a haemorrhage. Researchers and clinicians are hoping INTERACT2, an international, multicentre trial, will give a conclusive answer in the next couple of years.

A team in Manchester is weighing the risks and benefits of blocking inflammation in haemorrhagic stroke. Inflammation damages brain tissue but in haemorrhage it may also be beneficial. Blood and some of its breakdown products, like iron, are toxic to the brain around it. The immune cells within the brain, the microglia, which are activated by inflammation, seem to have an important role in mopping up some of those toxic products.

The team, lead by Dr Adrian Parry Jones, is using imaging to establish the time course of inflammation in haemorragic stroke – at what stage it is beneficial, and when it becomes harmful. They will then explore the effects of blocking it with a drug called Interleukin1 receptor antagonist (IL1RA) in a rodent model.

At present, surgery is likely to offer the greatest hope for haemorrhage. Researchers and stroke specialists are awaiting the results of three big international surgical trials, all aiming to get the blood out of different parts of the brain. STICH2 is looking at whether patients with a haemorrhage near the surface of the brain will benefit from a craniotomy to drain the blood away. CLEARIII and MISTIE [PDF] are testing minimally invasive surgical techniques, supplemented with tPA, to drain blood from the ventricles and from the haematoma itself, respectively.

The field abounds with possibilities and interesting data, but there is as yet no conclusive evidence to tell clinicians what they should be doing as a matter of routine.

Time is brain

With all brain injuries, time is brain. The faster we can intervene, the more brain tissue we can save. One way of doing that would be to make diagnosis – ischaemia or haemorrhage – and clot-busting treatment (if applicable) sooner, from the hospital to the ambulance. To that end, researchers in Germany recently installed a scaled-down CT scanner in an ambulance. Patients with an ischaemic brain could then be started on tPA and taken directly to a stroke unit or, if the clot is large, to a catheter laboratory to have it broken down by a clot-retrieval device (routinely used in Germany). The trial was a success and researchers are currently applying for funds to do the same thing in the UK.

Another way forward would be to develop a treatment that would work for any kind of stroke, regardless of whether a patient has bleeding or a clot in the brain. Doctors routinely induce hypothermia to save tissue in partients with cardiac arrest, and now the EuroHYP-1 [PDF] trial aims to test the effect of inducing hypothermia in 1500 ischaemic stroke patients in 20 different European countries. The treatment could be started in an ambulance, by injecting a saline solution, and continued in hospital by pouring cold water through adhesive pads on the patient’s skin. If it does prove to improve the outcome for ischaemic stroke patients, Malcolm Macleod, one of the principal investigators, says, it could be trialled in haemorrhage.

It’s not straightforward, though. Inducing hypothermia is a fairly aggressive procedure that brings on violent shivering. It may also suppress the immune system, increasing the risk of pneumonia. If adopted as standard practice it would mean rethinking stroke units again – possibly installing intensive care units within them – and further reconfiguring services.

Hugh MacCann’s fEITER imaging system may also be suitable for ambulance-based diagnosis and monitoring. It’s lightweight and portable – and once the team have established whether the REG signal is related to changes in blood flow or in pressure within the brain, they may be able to use the signal to detect whether the stroke has been caused by a clot or haemorrhage respectively.


Another area teeming with unanswered questions is what happens after the emergency – the acute stroke – is over and whatever damage is done is done. Our understanding of how the brain reacts to stroke, how it reorganises and repairs itself with time, and how much potential there is to get additional recovery of function over and above what nature achieves, is still limited.

But researchers are starting to turn their attention to long-term recovery from stroke. Ongoing trials are exploring whether small molecules and drugs – such as the L-DOPA dopamine agonist drugs used in Parkinson’s disease – can help retrain the brain and enhance somebody’s ability to learn new things.

Another approach is to find ways of encouraging the growth of new nerve cells. Keith Muir is collaborating with an academic spin-out company, ReNeuron, who have developed a cell line from human fetal cells. They have already shown that injecting stem cells into the region of a rat brain that’s been damaged by a stroke gives the rat better recovery, and they are now testing the therapy in humans (PISCES). The cells have been implanted in five patients, who are now being followed up for up to 18 months.

Although most of the implanted cells won’t survive and will get disposed of by the body, the ones that do will churn out messengers, chemicals and factors to stimulate repair processes in the brain – and potentially form some new connections.

Stroke is an insidious enemy and we’ve still got a long way to go before we defeat it. But with so many ideas bubbling around and trials in the pipeline it looks as if the next few years might transform the field once again – as radically as tPA and stroke units have done. Establishing a proper network of research centres all contributing to clinical trials will be critical to that. It is, according to Keith Muir, the only way forward. “Anything else is just guessing.”


1. National Institute for Neurological Disorders and Stroke rt-PA Study Group. Tissue plasminogen activator for the treatment of acute ischaemic stroke. N Engl J Med 1995;333:1581–88.

2. B Indredavik et al. Benefit of a stroke unit: a randomized controlled trial. Stroke 1991;22:1026-1031.

3. Langhorne et al. Do stroke units save lives? Lancet. 1993 Aug 14;342(8868):395-8.

Wellcome Trust funding

Dr Stephen Payne is at the Oxford Centre of Excellence in Personalised Healthcare, jointly funded by the Wellcome Trust and the Engineering and Physical Sciences Research Council.

Professor Hugh McCann received a Wellcome Trust Translation Award to develop fEITER technology in 2005.

This article is part of the Wellcome Trust’s Focus on stroke, a series of articles, interviews and videos running throughout May 2012, which is the Stroke Association’s Action on Stroke Month.

For more information on stroke, visit the Stroke Association‘s site or call its helpline on 0303 303 3100. If you or someone with you is suspected of having a stroke, call the emergency services immediately.

5 Comments leave one →
  1. Kim Looi permalink
    16 May, 2012 12:37 am

    A welcome update on a subject area of health in present day.

  2. 20 May, 2012 12:38 pm

    Peter Sandercock writes the blog doesn’t mention IST-3. It is the largest ever trial of the clot-busting drug rt-PA; it includes 3035 patients, 1617 of whom are aged over 80. It will therefore provide unique evidence on the effects of thrombolysis in people aged over 80 who have been excluded from treatment trials in the past. The results will be published in the Lancet at 12.00 midday on 23rd May, to coincide with presentation at the European Stroke Conference in Lisbon. More details will be made available at you can also follow @IST_3 on twitter #strokethrombolysis.

  3. 31 May, 2012 4:43 pm

    The ist-3 trial results are now published in the lancet, along with an updated meta-analysis of all completed trials of rt-PA for stroke


  1. Focus on stroke: Research in rehab « Wellcome Trust Blog
  2. Focus on Stroke « Wellcome Trust Blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: