Evidence is growing that early mechanical removal of clots from blocked brain arteries may improve recovery after stroke

Restoring blood flow to the brain within a few hours of having a stroke improves a person’s chances of recovery. Early use of drugs to dissolve clots (thrombolysis) has been the mainstay of management in recent years and there has also been rapid development of devices to remove clots.

Using evidence from trials published in the last five years, this review found mechanical removal of clots was more effective than thrombolysis alone whereas earlier research of older techniques had suggested that mechanical removal was no more effective than thrombolysis. The use of mechanical devices requires expertise and equipment as well as rapid access to brain scanning soon after stroke.

Given that stroke is both common and causes disability for many people, even a small improvement in the proportion of people for whom blood flow is restored soon after stroke could greatly reduce overall suffering and care costs.

More information is needed on which clot removal device to use, which patients benefit the most, the optimal timing of treatment and the effect of using different thrombolysis drugs alongside surgery. The UK-based PISTE trial may help answer some of these questions in the next few years.

Why was this study needed?

There are about 152,000 strokes per year in the UK, of which 85% are due to clots in brain arteries and smaller blood vessels. Stroke is the fourth most common cause of death in the UK and half of all survivors are left with a disability. The direct and indirect costs of stroke in the UK to health and social care services, survivors and their families is between £3.7 and £9 billion per year. Early treatment within hours of the stroke reduces the likelihood of long-term disability and might help reduce these costs. While ‘clot busting’ drugs such as alteplase help some people, some trials have showed that mechanical removal of clots can give better ability to function by 90 days after the stroke. However, the results aren’t consistent, and the reviewers speculated this could be due to differences between older and newer devices and approaches. This review aimed to look at the most recent evidence.

What did this study do?

The study was a systematic review of eight multi-site randomised controlled trials, including a total of 2,049 participants. The trials were carried out in high-income countries between 2013 and 2015, so the research is relevant for the UK and up to date. The average age of participants was between 64 and 71. The trials compared the effectiveness and safety of mechanical clot removal with usual care, which included thrombolysis for suitable patients, in the 6-8 hours after a stroke caused by a clot blocking brain arteries, The main outcome was achieving a “good” score of 0-2 on a scale of disability and dependence 90 days after the stroke occurred, where 0 is perfect health and 6 is death. The review was well designed but there was large variation between the eight trials in the treatments used and how they defined severity of strokes or age groups of patients. There was also insufficient information to analyse the results according to device. This slightly reduces our confidence in the findings and the ability to practically apply the findings.

What did it find?

  • Pooled results from all eight trials showed that 43.2% of participants who had mechanical clot removal showed “good” functioning 90 days after the stroke compared with 30.9% of those who received usual care without mechanical clot removal (odds ratio [OR] 1.75, 95% confidence interval [CI] 1.20-2.54). The number of people who need to be treated mechanically instead of usual care for one extra person to attain the “good” functioning category was eight (NNT=8).
  • A slightly stronger effect was seen in the most recent five trials, published in 2015, in which 46.1% of participants undergoing mechanical clot removal gained good functioning, whereas only 26.4% of control patients did so (NNT=5).
  • Six trials used imaging to confirm the presence of a clot in a brain artery. When a clot was confirmed before treatment, mechanical clot removal gave a better chance of achieving good functioning by 90 days, OR 2.00, 95% CI 1.48-2.71.
  • Those under 70, over 70, receiving intravenous thrombolysis, with worse strokes, and patients with moderate strokes all had increased odds of a good outcome.
  • Mortality was slightly less common in the intervention groups (15.36%) than the control groups (18.24%), but these findings were not statistically significant.

What does current guidance say on this issue?

NICE guidance, published in 2013, on mechanical removal of clots for stroke says the effectiveness is unproven, and there are risks of serious complications, but this does not include evidence from the more recent trials.

Balancing these risks with the poor prognosis of many patients with stroke, the 2013 guidance recommends mechanical clot retrieval should only be used in the context of research for patients with acute ischaemic stroke for whom thrombolysis is suitable. For those where thrombolysis is not suitable there should be special arrangements for clinical governance, consent and audit or research. NICE recommends that any further research should choose participants who meet the same suitability criteria for thrombolysis as the on-going PISTE trial. These criteria include the participant being independent before the trial, having a known clot in particular brain arteries, being treatable within 6 hours of stroke, and being able to return for follow up three months later.

An international statement on clot retrieval (2014) recognised that trials of later types of mechanical clot removal device, called stent retrievers, showed that they may be more effective than using thrombolysis alone, but that further research was needed into the impact of using different devices.

What are the implications?

Current stroke treatments still leave many people with long term disability. Because stroke is so common and can be so disabling, many people could be helped if the chances of recovery improved for even a small proportion of those treated. It may be too early to add mechanical clot removal to standard treatment for the most common type of strokes, despite its promise in this review. There are considerable implications for NHS resources as these devices are costly and require rapid access to sophisticated brain imaging. The best type of device to use, the choice of thrombolytic drug to use alongside (if any), and the best time window to give clot removing treatment are still to be established. Similarly, the explanation for why more recent studies showed greater effectiveness than those from more than five years ago needs confirmation. The working hypothesis is that newer devices are better than the old. The ongoing, UK-based PISTE trial will help answer some of these questions. It is expected to run until at least 2018.

NICE guidance from 2013 says that mechanical clot retrieval should only be used in trials. Implementing the procedure as standard care in the UK could increase costs of training and treatment, and may require further restructuring of stroke services. On the positive side, it might save costs by reducing disability and long term care. Economic analysis, a training plan, as well as more information about what works for whom, would inform any decisions to adopt this promising technology more widely.


Yarbrough CK, Ong CJ, Beyer AB, et al. Endovascular thrombectomy for anterior circulation stroke: systematic review and meta-analysis. Stroke. 2015;46(11):3177-83.


Consensus statement on mechanical thrombectomy in acute ischemic stroke – ESO-Karolinska Stroke Update 2014 in collaboration with ESMINT and ESNR. ESO-Karolinska Stroke Update Conference, Stockholm, 16-18 November 2014.

NICE. Mechanical clot retrieval for treating acute ischaemic stroke. IPG458. London: National Institute for Health and Care Excellence; 2013.

Pragmatic Ischaemic Stroke Thrombectomy Evaluation (PISTE): protocol and criteria [internet]. University of Glasgow.

Stroke Association. State of the nation - stroke statistics January 2015. London: Stroke Association; 2015.

van Swieten J, Koudstaal P, Visser M, et al. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988;19(5):604–7.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre


A range of different devices and approaches have been developed to remove clots which are blocking brain arteries and causing stroke.

So called ‘first generation’ devices include using catheters to dislodge the clot, removing it using suction, and wrapping a coil around the clot before removing it using a catheter.

‘Second generation’ devices are called stent retrievers. These are telescopic devices that capture the clot on outer surface of the stent at the same time as restoring flow to the artery.

To assess participants’ functioning after stroke, this study used the modified Rankin Scale. Health professionals interview the patient using a standard questionnaire, using the results to score their functioning on a scale from 0-6, where:

0 = No symptoms at all

1 = No significant disability despite symptoms; able to carry out all usual duties and activities

2 = Slight disability; unable to carry out all previous activities, but able to look after own affairs without assistance

3 = Moderate disability; requiring some help, but able to walk without assistance

4 = Moderately severe disability; unable to walk without assistance and unable to attend to own bodily needs without assistance

5 = Severe disability; bedridden, incontinent and requiring constant nursing care and attention

6 = Dead



Expert commentary

This systematic review and meta-analysis summarises the results of eight randomised controlled trials which evaluated endovascular thombectomy for acute ischaemic stroke. The results, which include five positive trials published in 2015, clearly show that endovascular thrombectomy undertaken within six hours of symptom onset is a very effective treatment for carefully selected patients (NNT is 5). There is on-going debate about the proportion of patients with acute ischaemic stroke who could be eligible to receive this treatment.

Endovascular thrombectomy is a valuable addition to hyperacute stroke care. UK stroke services need to address how this time dependent intervention can be provided 24/7 and service reconfiguration is likely to be needed in some areas.

Professor Helen Rodgers, Clinical Professor of Stroke Care, Stroke Research Group, Institute of Neuroscience and Newcastle University Institute for Ageing; Incoming President of the British Association of Stroke Physicians